This page intentionally left blank

CENTRAL PAIN SYNDROME

Central Pain Syndrome (CPS) is a neurological condition caused by damage
specifically to the central nervous system
brain, brainstem, or spinal cord.
This is the only up-to-date book available on the clinical aspects (including
diagnosis and therapy) of CPS management. The authors have developed
a complete reference source on central pain, which includes background
material, the pathophysiology of the syndrome, and diagnostic and therapeutic
information. The syndrome has been a medical mystery for 100 years with no
effective cure; this book turns the concept of incurability of central pain on its
head, providing a rational approach to therapy based on a scientific theory.
Sergio Canavero set out to become a functional neurosurgeon after reading
Scientific Americans special issue on the brain in September 1979. He graduated
magna cum laude and went on to gain FMGEMS certification. Finding psychosurgery impossible to pursue in Italy, he moved on to the field of pain and
movement disorders, introducing extradural cortical stimulation for Parkinsons
disease and stroke rehabilitation. His lifetime focus is on the nature of consciousness. With Bonicalzi, he founded the Turin Advanced Neuromodulation
Group, a think tank focusing on the advancement of neuromodulation. His
secondary focus is on women, a subject he discussed in his book Donne Scoperte
(Women Unveiled, 2005), which attracted media interest.
Vincenzo Bonicalzi graduated magna cum laude and took up a career in
anesthesiology, pain therapy, and intensive care. He is a senior staff member
at the Department of Neurosciences at the third-largest medical facility in
Italy. He is an enthusiast for medical statistics and evidence-based medicine.
With Canavero, he has explored the pitfalls of modern neurointensive care.

CENTRAL PAIN SYNDROME

PATHOPHYSIOLOGY, DIAGNOSIS
AND MANAGEMENT
Sergio Canavero, MD (US FMGEMS)
Turin Advanced Neuromodulation Group (TANG), Turin, Italy

Vincenzo Bonicalzi, MD
Turin Advanced Neuromodulation Group (TANG), Turin, Italy

CAMBRIDGE UNIVERSITY PRESS

Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, So Paulo

Cambridge University Press
The Edinburgh Building, Cambridge CB2 8RU, UK
Published in the United States of America by Cambridge University Press, New York
www.cambridge.org
Information on this title: www.cambridge.org/9780521866927
A. Burns and A. Wellings 2007
This publication is in copyright. Subject to statutory exception and to the provision of
relevant collective licensing agreements, no reproduction of any part may take place
without the written permission of Cambridge University Press.
First published in print format 2006
eBook (EBL)
ISBN-13 978-0-511-29525-6
ISBN-10 0-511-29525-1
eBook (EBL)
ISBN-13
ISBN-10

hardback
978-0-521-86692-7
hardback
0-521-86692-8

Cambridge University Press has no responsibility for the persistence or accuracy of urls
for external or third-party internet websites referred to in this publication, and does not
guarantee that any content on such websites is, or will remain, accurate or appropriate.

To my parents, for their unfailing support

and
to Serena, Marco and Francesca, wishing them a world without pain

To my parents, source of my past

and
to Cecilia, path of my future

CONTENTS

List of Tables
Preface (or, the Story of an Idea)
1

page xi
xiii

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Definitions
History

1
3

Central Pain of Brain Origin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

Epidemiology and clinical features
1. Lesions causing CP and location
2. Incidence and prevalence
3. Age of onset of CPSP
4. Sex distribution of CPSP
5. Time to pain onset
6. Side of the lesions
7. Size of the lesion and CPSP
8. Pain distribution
9. Quality of pain
10. Intensity of pain
11. Components
12. Evoked pains
13. Somatosensory findings
14. Sympathetic and other signs and symptoms
Special considerations
1. Sensory epilepsy
2. Parkinsons disease (PD)
3. Iatrogenic lesions

Central Pain of Cord Origin

1. Lesions causing CCP
2. Incidence and prevalence

9
9
22
28
28
28
40
40
40
47
49
49
54
56
99
100
100
101
102

. . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
113
113
vii

viii

Contents

3. Age of onset and sex distribution

4. Time to pain onset
5. Level of lesion
6. Distribution of pain
7. Quality
8. Intensity
9. Components
10. Evoked pains
11. Somatosensory deficits
12. Sympathetic and other signs and symptoms
13. Course
Special considerations
1. Syringomyelia
2. Multiple sclerosis (MS)
3. Spinal epilepsy

116
116
117
118
119
120
120
120
121
128
128
129
129
130
132

Diagnosing Central Pain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133

Drug Therapy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138

Oral and parenteral drug therapy
1. General comments
GABA drugs
Antiglutamatergic agents
Sodium channel blockers
Antidepressants
Opioids, naloxone and cannabinoids
Miscellaneous agents
Combinations
2. Drug aggravation of CP

139
139
139
177
178
179
180
181
181
182

Neuromodulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
Electrical
1. Extra- and subdural motor (MCS) and sensory cortex
stimulation (PCS)
Mechanism of action
2. Deep brain stimulation (DBS)
Mechanism of action
a. PAG-PVG
b. Vc
c. Other areas
Efficacy
a. Vc DBS
b. PAG-PVG DBS
3. Spinal cord stimulation
Mechanism of action
Efficacy

183
184
195
199
199
199
206
208
208
209
209
211
211
211

Contents

ix

4. Transcutaneous electrical nerve stimulation (TENS)

Mechanism of action
Efficacy
5. Gasserian ganglion stimulation
6. Vagal nerve stimulation
7. Electroconvulsive therapy
Mechanism of action
Efficacy
8. Conclusions
Chemical
Prevention
Alternative approaches
A future?
7

Pathophysiology: Human Data

218
218
220
221
221
221
221
223
223
223
232
232
235

. . . . . . . . . . . . . . . . . . . . . . . . . . 237

Functional imaging studies

1. Studies assessing the spontaneous resting component
and its modulation
2. Studies assessing the evoked components
3. Studies assessing biochemical changes
Interpretation
Neurophysiology
1. Human microrecording/stimulation studies
Findings in BCP
Findings in CCP
Spinal recordings
Cerebral recordings
Findings in mixed series
Interpretation
Evoked potentials studies
Interpretation
Results of neuroablation
Interpretation
a. Pre- and post-central gyrectomy
b. Frontal (psychiatric) surgery (lobotomy, topectomy,
cingulectomy/cingulotomy, leukotomy)
c. Hypothalamotomy and hypophysectomy
d. Thalamotomies
e. Mesencephalotomies
f. Other brainstem procedures
g. Anterolateral cordotomies (spinothalamic tractotomies)
h. Posterior cordotomies and commissurotomies
i. Cordectomies
j. Dorsal root entry zone (DREZ) lesions
k. Spinal rhizotomies

239
240
243
247
249
254
254
254
256
256
257
259
264
265
267
267
267
267
293
294
294
297
297
298
299
299
299
300

Contents

l. Sympathectomies and sympathetic blocks

Reports of sudden disappearance of CP
8

300
300

Piecing Together the Evidence . . . . . . . . . . . . . . . . . . . . . . . . . . 307

Exploring the theoretical framework
Triggering persistent oscillation
What releases CP?
Dissecting the role of GABA
Which role for neuroplasticity?
Bilaterality of central pain
Lessons from cord central pain
The genesis of allodynia

References
Index

310
311
315
321
324
327
330
336
339
379

LIST OF TABLES

1.1
2.1
2.2

2.3
2.4
2.5
2.6
2.7
2.8
2.9
2.10
2.10a
2.10b
2.11
2.12
2.13
3.1
3.2
3.3
3.4
3.5
3.6
4.1
5.1
5.2

Historic highlights of central pain in the western literature

(from Garcin 1937; DeAjuraguerra 1937)
Lesions causing CP and their location
Incidence and prevalence (only best evidence studies included;
excluded ones can be found in bibliographies attached
to cited papers)
Age of onset (CPSP only)
Sex distribution (CPSP only)
Timespan between stroke injury and CP onset
Side of lesion
Size of the lesion
Pain quality
Central neurogenic pruritus due to brain lesions (excluding MS)
Clinical features (selected studies)
Somatosensory troubles in Head and Holmes cases of
thalamic pain
Review of somatosensory abnormalities in Head and Holmes
published cases
Somatosensory troubles in the review of De Ajuriaguerra (1937)
Somatosensory findings in selected series
Iatrogenic lesions originating central pain (selected studies)
Causes of cord central pain (compiled from a complete survey
of the literature and personal observations)
Distribution of causes of CCP
IASP classification of SCI pains (adapted from Siddall et al. 2001)
Incidence and prevalence
Clinical features (representative series; Defrin et al. 2001)
Somatosensory findings in selected series
Examination protocol for central pain (IASP)
Controlled studies of oral drugs
Controlled studies of parenteral drugs

3
10

23
29
31
36
41
45
48
50
57
64
68
69
70
104
114
114
115
117
121
122
136
140
149
xi

xii

List of Tables

5.3
5.4
6.1
6.2
6.3
6.4
6.5
6.6
6.7
6.8
7.1

Uncontrolled studies: oral drugs

Uncontrolled studies: parenteral drugs
Motor cortex stimulation (MCS) (arranged by reporting groups)
Deep brain stimulation (DBS) arranged by reporting groups
Spinal cord stimulation (dorsal column stimulation) (SCS)
Transcutaneous electrical nerve stimulation (TENS)
Gasserian ganglion stimulation
Electroconvulsive therapy (ECT)
Chemical neuromodulation (spinal intrathecal (IT) or
epidural (EPI) infusion)
Treatment of central pain: the TANG guidelines
Results of neuroablative procedures

160
171
185
200
212
219
221
222
224
235
268

PREFACE (OR, THE STORY OF AN IDEA)

The man with a new idea is a crank
until the idea succeeds. (Mark Twain)

The story of this book goes back 15 enthusiastic years. At the end of 1991, S.C., at the
time 26, was asked by C.A. Pagni, one of the past mavens of the field, to take up
central pain. S.C. was back from a semester as an intern at Lyon (France)
neurosurgical hospital. A dedicated bookworm, he often skipped the operating
theater in favor of the local well-stocked library. In that year a paper was published
by two US neurobiologists, espousing the idea of consciousness arising from
corticothalamic reverberation: this paper drew his attention, as he was entertaining
a different opinion as to how consciousness arises. At the beginning of 1992 he came
across a paper written by two US neurologists, describing a case of central post-stroke
pain abolished by a further stroke: the authors were at a loss to explain the reason.
Discoveries sometimes happen when two apparently distant facts suddenly fit
together to explain a previously puzzling observation. And so it was. During a girlhunting bike trip at Turins best-known park, a sunny springtime afternoon, the
realization came thundering in. Within a short time, a name was found and so the
dynamic reverberation theory of central pain was born. It was first announced in
a paper published in the February 1993 issue of Neurosurgery and then in Medical
Hypotheses in 1994.
In May 1992 Pagni introduced Dr. Bonicalzi, a neuroanesthesiologist and pain
therapist, to S.C. Over the following years, the combined effort led to further
evidence in favor of the theory, in particular a neurochemical foundation based
on the discovery that propofol, a recently introduced intravenous anesthetic, could
quench central pain at nonanesthetic doses (September 1992). The idea of using
propofol at such dosage came from reading a paper by Swiss authors describing
its use in central pruritus. The similitude between central pain and pruritus, at
the time not clearly delineated in the literature, was the driving reason. In 1988
Tsubokawa in Japan introduced cortical stimulation for central pain: it was truly
ad hoc, as cortex plays a major role in the theory. Happily, since 1991, the cortex
has gone through a renaissance in pain research, although neurosurgical work
already pointed in that direction. We soon combined three lines of research
drug

xiii

xiv

Preface

dissection, neuroimaging and cortical stimulation data
in our effort to tease out
the mechanism subserving central pain.
Central pain as a scientific concept was the product of an inquisitive mind, that of
Dr. L. Edinger, a neurologist working in Frankfurt-am-Main, Germany, at the end of
the 1800s. Despite being recognized by early-twentieth-century neurologists as
the initiator of the idea of centrally arising pains, this recognition soon faded,
shadowed by Dejerine and Roussy and their thalamic syndrome. At the beginning of
the twenty-first century, due credit must go to the physician who deserved it in the
first place, namely Dr. Edinger.
For a century, central pain has remained neglected among pain syndromes, both
for a lack of pathophysiological understanding and a purported rarity thereof.
Far from it! Recent estimates make it no rarer than Parkinsons disease, which,
however, commands a huge literature. Worse yet, the treatment of central pain has
only progressed over the past 15 years or so and much of the new acquisitions have
not yet reached the pain therapist in a rational fashion.
As we set out to write this book, we decided to review the entire field and not only
expound the dynamic reverberation theory, which, as we hope to show, may well
represent the end of central pain. It has truly been a sweatshop work as we
perused hundreds of papers and dusted off local medical libraries in search of obscure
and less obscure papers in many languages, as true detectives. We drew out single
cases lost in a mare magnum of unrelated data and in the process gave new meaning
to long-overlooked reports. We also realized that some bad science mars the field,
and this is properly addressed.
The result is
hopefully
the most complete reference source on central pain
in the past 70 years or so. The reader should finish the book with a sound
understanding of what central pain is and how it should be treated. The majority
of descriptive material has been tabulated, so that reading will flow easily. We hope
this will be of help to the millions who suffer from central pain.
Special thanks go to the unsung heroes at the National Library of Medicine in
Washington, DC, whose monumental efforts made our toil (and those of thousands
of researchers around the world) less fatiguing. Thanks also to the guys behind
Microsoft Word, which made the tabulations easy as pie. Also, due recognition
must go to the Cambridge staff who have been supervising this project over the
past two years, especially Nat Russo, Cathy Felgar, and Jennifer Percy and the
people at Keyword, above all Andy Baxter and Andrew Bacon for the excellent
editorial work.
Sergio Canavero, Vincenzo Bonicalzi
Turin, May 2006

INTRODUCTION

Frau R. Suicidium. (Edinger 1891)

DEFINITIONS
Ever since Dejerine and Roussys description of central pain (CP) after thalamic
stroke in 1906, thalamic pain (itself part of the thalamic syndrome) has remained the
best-known form of CP and it has often
misleadingly
been used for all kinds
of CP. Since CP is due to extrathalamic lesions in the majority of patients, this term
should be discarded in favor of the terms central pain of brain
brainstem or cord
origin (BCP and CCP). Other terms that are now obsolete and should be discarded
include pseudothalamic pain (i.e., CP caused by extrathalamic lesions) and anesthesia
dolorosa, when this refers to CP in an anesthetic region caused by neurosurgical
lesions. If a stroke at whatever level is the cause of CP, the term central post-stroke
pain (CPSP) is used. Even though some clinical features are similar, peripheral
neuropathic pain (PNP) is not CP.
CP is akin to central dysesthesias/paresthesias (CD) and central neurogenic
pruritus (CNP): actually, these are facets of a same disturbance of sensory processing
following central nervous system (CNS) lesions. Dysesthesias and paresthesias differ
from pain in their being abnormal unpleasant and non-unpleasant sensations with a
nonpainful quality. While contributing to suffering, they can also be found in PNP.
Dysesthetic pain used as a synonym of CP must also be abandoned.
Since 1978 there has been a tendency to combine CP and PNP under the general
rubric of deafferentation pain on account of shared clinical features, both being due
to a decrease in afferent input into the CNS and consequent sensitization (see Tasker
2001). Deafferentation pain has never been included in the taxonomy published by
the International Association for the Study of Pain (IASP) (Merksey and Bogduk
1994), and actually indiscriminate lumping of all neuropathic pains under this term

Central Pain Syndrome

has created much confusion and even contradictions, often hindering assessment
of therapeutic strategies for single disease entities. The term neural injury pain
should also be discarded. In 1990 a consensus group (Devor et al. 1991) concluded
that: The term deafferentation pain as presently used is misleading and should
perhaps be abandoned altogether for purposes of clinical diagnosis.
Virtually all kinds of slowly or rapidly developing disease processes affecting the
spino- and quintothalamic pathways (STT/QST), i.e., the pathways that are most
important for the sensibility of pain and temperature, at any level from the dorsal
horn/sensory trigeminal nucleus to the parietal cortex, can lead to CP/CD/CNP.
These do not depend on continuous receptor activation.
The IASP defines CP as pain initiated or caused by a primary lesion or
dysfunction of the central nervous system (Merksey and Bogduk 1994), i.e., of the
spinal cord, brainstem or cerebral hemispheres. This definition is too extensive, as it
includes pain associated with motor disorders (Parkinsons disease and dystonia)
and painful fits, which
although being CNS disorders
are not strictly CP:
impairment of spinothalamocortical conduction, a cardinal finding of CP, is not
seen in these conditions. However, there are cases of bona fide CP without clinical
or electrophysiological signs of such impairment. We propose that CP/CD/CNP be
considered only spontaneous, constant and/or evoked pain, dysesthesia or pruritus
initiated by a CNS lesion impinging on or interfering with the spinothalamoparietal
path. Since CP appears to be the most frequent of these three conditions, we will
generally refer to CP throughout the text. Parkinsons disease, epileptic pains and
perhaps other diseases with a painful CP-like component should be classified as
central pain-allied conditions (CPAC).
Once thought an uncommon neurological curiosity, CP is an important and
underrecognized condition. CP produces immense suffering (a great burden), even
when intensity is low: its generally very unpleasant and irritating, largely constant
character makes it incomprehensible by almost all sufferers. Patients can be
completely disabled and CP may be so devastating as to override any other disability
in the chronic stage. By dominating the sensorium, interfering with the thought
processes and undermining the morale, CP frequently alters mood, intellect and
behavior with deterioration of personality, depression and neurotic tendencies,
interfering with rehabilitation, and impairing daily activities and quality of life. Many
patients with severe persistent pain undergo a progressive physical deterioration
caused by disturbance of sleep and appetite, a restriction in physical and daily
activities, and often become addicted to medications, all of which contribute to
general fatigue, increased irritability and decreased libido and sexual activity. The
social effects are equally devastating: many patients have progressively greater problems with their families and friends, reduce their social interactions and activities
and are unable to work (Widar et al. 2004). There are hints that chronic pain may
suppress the immune system and even alter insulin sensitivity. Some patients with
severe persistent pain become so discouraged and desperate that they commit suicide, and usually not because of depression. Last, but not least, CP financially
burdens both society and patients. Thus, it represents a true challenge.

Introduction

HISTORY
Those who cannot remember the past are condemned to repeat it. (G. Santayana)

Cases of CP following brain or cord damage have most certainly been observed since
antiquity, but never understood as such. We have to wait until the nineteenth century
for published descriptions of what we now understand to be CP (Table 1.1) in
western medicine (there appear to be reports of what is most likely CP in ancient
Chinese medicine, this being the result of a deficiency of the Qi and attendant blood
stasis, in turn depriving the nourishing of meridians and tendons; see Kuong 1984).
TABLE 1.1. Historic highlights of central pain in the western literature (from

Garcin 1937; DeAjuriaguerra 1937)*

Marcet (1811)

Describes pain after bulbar lesions

Fodera (1822)

Describes pain after spinal hemisection

Brown-Sequard (1850)

Describes the syndrome named after him;

confirms previous description of hyperesthesia
below lesion level on the plegic side
Descriptions of pain after spinal trauma during
the U.S. Civil War
Further describes pain after bulbar lesions

1860
70s
Marot (1875)
Nothnagel (1879)

Page (1883)

First precise description of constant pain following

tumors of the pons Varolii (mentioned by other
authors) and other sites
Describes pain in spinal cord injury patients

Edinger (1891)

Birth of the concept of central pain

Hardford (1891)

Describes pain of cortical origin

Mann (1892)

Matches CP to infarctions of medulla at

nucleus ambiguus level
Describes syringomyelic pain

Gilles de la Tourette (1889)

Wallenberg (1895)

Link (1899)

Describes the syndrome named after him;

insisted on facial pains; ascribed it to
PICA embolism (verified autoptically in 1901)
Describes CP as resulting from parietal stroke
(autopsy confirmed)
Describes CP as resulting from pontobulbar lesions

Dejerine and Roussy (1906)

Describe the syndrome named after them

Holmes (1919)

Typical thalamic pain observed in spinal cord

injured patients (World War I soldiers)
Autoptic confirmation that CP may arise
without thalamic involvement

Reichenberg (1897)

Souques (1910), Guillain and Bertrand,

Davison and Schick, Schuster,
Wilson, Parker (1920s
30s)

* A great many authors described CP, but dates are not available through the two cited reviews: these authors include
Halische, Joly, Duchek, Biernacki, Oppenheim, Bechterew (pre-1900), Barre`, Elsberg (cordonal pain), Foerster (dorsal
horn pain), Vulpian, Gowers, Gerhardt (recognized CP in multiple sclerosis), Schlesinger, Lhermitte and DeMassaryBonhomme (hematobulbia), Mills, Mattirolo, Hanser, and many others.

Central Pain Syndrome

However, the possibility of centrally arising pains was simply dismissed by most
authorities.
It was only in 1891 that Edinger, a German neurologist, challenging the prevailing
opinion of the day, and avec une rare sagacite (with rare sagacity; Garcin 1937),
introduced the concept of centrally arising pains. In his landmark paper Giebt es
central entstehende Schmerzen? Mittheilung eines Falles von Haemorrhagie in den
Nucleus externus Thalami optici und in das Pulvinar, dessen wesentliche Symptome in
Hyperaesthesie und furchtbaren Schmerzen in der gekreuzten Seite, ausserdem in
Hemiathetose und Hemianopsie bestanden haben (Are there centrally arising pains?
Description of a case of bleeding in the nucleus externus thalami optici and in the
pulvinar, whose essential symptom consisted in hyperesthesia and terrible pains in
the contralateral side, besides hemiathetosis and hemianopsia), he remarked how
only a few cases of pains associated with damage of the brain, brainstem and spinal
cord were on record (Die Durchsicht der Literatur nach aehnlichen Beobachtungen
hat nur wenig ergeben
A literature review of similar cases has borne little fruit),
but that other reasons were adduced to explain them (generally peripheral nerve
causes or muscle spasms). One of the few well investigated cases was that of
Greiff (1883), concerning a 74-year-old woman who developed Hyperaesthesie und
reissenden Schmerzen im linkem Arm, geringgradiger im linkem Beine (hyperesthesia
and tearing pains in the left arm and of lesser intensity in the left leg) as a
consequence of several strokes and which lasted for two months until death. At
autopsy, two areas of thalamic softening were found, one of which was in what
appears to be ventrocaudalis (Vc). Greiff commented on vasomotor disturbances
as a possible cause of pain. According to Edinger Vielleicht giebt es auch corticale
Schmerzen (perhaps there are also cortical pains), and he cited as evidence
. . . schmerzhaften Aura bei epileptischen, abnorme Sensationen bei Rindenherden und
Reizerscheinungen im Bereich des Opticus bei Affectionen des Hinterhaupts-lappens
(. . . painful aura in epileptics, abnormal sensations in cortical foci and signs of
excitation in the territory of the opticus following diseases of the occipital lobe).
Edinger reported on einen Krankheitsfall . . . in dem als Ursache ganz furchtbaren
Schmerzen post mortem ein Herd gefunden wurde, der dicht an die sensorische Faserung
grenzend im Thalamus lag. Der Fall erscheint dadurch besonders beweiskraftig fuer die
Existenz centraler Schmerzen, weil die Hyperaesthesie und die Schmerzen sofort nach
dem Insulte und monatelang vor einer spaeter auftretenden Hemichorea sich zeigten
(a patient . . . in whom the origin of truly terrible pains was at autopsy a lesion that
impinged on the fibers abutting the thalamus. This case is thus especially convincing
evidence for the existence of central pains, as the hyperesthesia and the pains
showed immediately after the insult and months before a later arising hemichorea).
The patient was Frau R. (Mrs. R.), aged 48, who developed heftige Schmerzen und
deutliche Hyperaesthesie in den gelaehmten Gliedern (violent pains and clear-cut
hyperesthesia in the paretic limbs: right arm and leg); Wegen der furchtbaren
Schmerzen Suicidium 1888 (due to the terrible pains, suicide 1888). This woman
developed an intense tactile allodynia for all stimuli bar minimal, which hindered all
home and personal activities (e.g., dressing) and made her cry; also Laues wasser
wurde als sehr heiss, kaltes als unertraeglich schmerzend (lukewarm water was felt
as very hot, and cold water as intolerably painful) in both limbs. Very high doses

Introduction

of Morphium were basically ineffective. This patients pain reached intolerable

peaks, but sometimes could be tolerated for a few hours or at most half a day before
shooting up again. In this patient, Vasomotorische Stoerungen, wie sie in dem
Lauenstein (D.Arch.f.klin.Med. Bd.XX.u.A.)schen . . . Falle bestanden haben, sind nicht
zur Beobachtung gekommen (vasomotor disturbances, as present in Lauensteins
case, were nowhere to be observed). At autopsy, Der Herd im Gehirn nimmt also den
dorsalen Theil des Nucleus externus thalami und einen Theil des Pulvinar ein, er
erstreckt sich lateral vom Pulvinar fuer 1 mm in den hintersten Theil der inneren Kapsel
hinein. Der Faserausfall, der dort in Betracht kommt, ist sehr gering. (The brain lesion
involved the dorsal portion of the nucleus externus thalami and a portion of the
pulvinar, extending laterally from pulvinar for 1 mm into the most posterior part of
the inner capsule. The loss of fibers, that can be observed at this point, is minimal.)
Thus, in Greiffs and Edingers patients lesions were respectively found at autopsy
in right thalamic nucleus internus and ventral thalamus and in thalamic nucleus
externus and pulvinar.
Edinger should be given the credit as the one who introduced the concept of
CP to neurology, as he wrote: Man kommt zum Schlusse, dass hier wahrscheinlich
durch directen Contact der sensorischen Kapselbahn mit erkranktem Gewebe die
Hyperaesthesie und die Schmerzen in der gekreuzten Koerperhaelfte erzeugt worden
sind (one concludes that here both the hyperesthesia and the pains in the crossed
half of the body have been likely caused by direct contact of injured tissue with the
sensory path coursing in the internal capsule), actually being the first to propose
an irritative theory of CP. Incidentally, he stressed the importance of the internal
capsule, a forerunner of our theory (Canavero 1994).
One year later, Mann (1892), another German neurologist, concluded, in Edingers
wake, that CP can be also observed outside the thalamus, namely in the medulla
oblongata, thus antedating Wallenbergs classic description (autopsy of this patient
performed in 1912 confirmed Manns clinical diagnosis and the involvement of the
spinothalamic tract). Thereafter, an explosion of reports ensued. In the first decade of
the twentieth century, Dejerine and Roussy (1906) described six cases of what they
called Syndrome thalamique, whose signs and symptoms were summarized by
Roussy (1906) in his thesis:
1) slight hemiparesis usually without contracture and rapidly regressive;
2) persistent superficial hemianesthesia of an organic character which can in some
cases be replaced by cutaneous hyperesthesia, but always accompanied by
marked and persistent disturbances of deep sensations;
3) mild hemiataxia and more or less complete astereognosis.
To these principal and constant symptoms are ordinairement (ordinarily) added:
1) severe, persistent, paroxysmal, often intolerable pain on the hemiparetic side
unyielding to any analgesic treatment;
2) choreoathetotic movements in the limbs on the paralyzed side.
On the basis of an autopsy study of three cases, they concluded that the lesion
is localized to the external, posterior and inferior region of the thalamus (thus
including the main sensory nucleus Ventrocaudalis, or Vc), impinges on the median

Central Pain Syndrome

nuclei and, to a lesser extent, involves a part of the posterior limb of the internal
capsule. Certainly, the complete syndrome is very rare. In their original paper on the
syndrome thalamique, Dejerine and Roussy evaluated microscopically the thalamic
lesion responsible for the syndrome. In their first case they noted a lesion in the
posterior thalamus, involving both external and internal nuclei and the internal
capsule. The lesion impinged more diffusely on the external thalamic nucleus. In
their second case the lesion again impinged more on the external thalamic nucleus,
but they also noted the lesion of the internal and median nuclei, internal capsule and
pulvinar. The lesion also impinged on the posterior pulvinar. In their third case a less
extended lesion was noted, impinging on the posterior part of the thalamic external
nucleus, the internal and median nuclei, the posterior part of the internal capsule and
part of the lenticular nucleus. They concluded that the thalamic syndrome follows a
lesion of the postero-external part of the external thalamic nucleus, impinging also
on part of median and internal thalamic nuclei and on the near part of the internal
capsule.
A few years later, Head and Holmes (1911), on the basis of personal and literature
autoptic evidence, concluded that thalamic pain depends on the destruction of
the posterior part of the external thalamic nucleus. In their book-size article, they
provide the best and first quantitative description ever of somatosensory alterations
in CP patients (Chapter 2).
During World War I several observations on thalamic pains associated with
spinal cord war lesions were published, as previously done
but only descriptively

during the U.S. Civil War in the 1860s.
The term central pain was first used in the English literature by Behan (1914).
In 1933 Hoffman reported a tiny lesion in the most basal part of the Vc, where
spinothalamic fibers end (Hasslers Vcpc). This is probably the smallest reported
lesion causing CP.
In the 1930s three major reviews on CP were published (De Ajuriaguerra 1937;
Garcin 1937; Riddoch 1938). Here, the interested reader will find an unparalleled
review of the literature of the nineteenth and early twentieth centuries, plus
unsurpassed descriptions of CP, whose ignorant neglect (admittedly also out of
language barriers) on the part of modern investigators is responsible for several rediscoveries. Nothing new has been basically added to the clinical literature since.
Riddoch gave this definition: By central pain is meant spontaneous pain and painful overreaction to objective stimulation resulting from lesions confined to the substance of the central nervous system including dysaesthesiae of a disagreeable kind.
It was clear how thalamic pains could follow a lesion of the lateral thalamic area,
in the territories of the lenticulo-optic, thalamo-geniculate and thalamo-perforating
arteries, but also of the cortex (rarely), internal capsule, medulla oblongata and less
frequently the pons (no mesencephalic lesions were on record) and the spinal cord
(not infrequently; particularly following injury and syringomyelia). Thermoalgesic
sensory loss and somatotopographical constraints were clearly delineated. However,
De Ajuriaguerra, based on a patient with a thalamic lesion and CP without sensory
derangement described by Lhermitte, concluded against a role of the sensory relay
nuclei in the genesis of CP (actually that patient had minimal sensory loss and loss of
cells and fibers also included Vc).

Introduction

The most frequent cause of CP appeared to be vascular at all levels, except

the brainstem, where tumors, tuberculomas, multiple sclerosis, syringobulbia and
hematobulbia contributed; Millss 1908 patient suffered mostly central paresthesias.
Epileptic pains were also considered CP.
Unfortunately, over the years, despite ample evidence that other lesions can cause
CP as well, the term thalamic syndrome became synonymous with CP, despite it being
clear to many that it was not so.
In 1969 Cassinari and Pagni, in their monograph Central Pain: A Neurosurgical
Survey, wrote: the conclusions of the various workers who have tried . . . to identify
the structure in which lesions are responsible for the onset of central pain sometimes
conflict. The divergence of opinion is fairly easily explained by the fact that
spontaneous lesions are usually extensive, difficult to define, often plurifocal, and
affect several systems with different functions. By studying iatrogenic pure lesions
(which they equated to experimental lesions) giving rise to CP, they reached the
conclusion that the essential lesion was damage to the pain-conveying spinothalamocortical tract. Also, they observed how operations that interrupt the central pain
pathways in order to allay pain may themselves originate CP (sometimes more severe
than the pain that led to the operation), an occurrence practically impossible to
foresee. However, the genesis of CP remained an enigma.
Thereafter, the subject received little additional attention (the hidden disorder:
Schott 1996). CP remained a neglected field among most medical educators and also
among neurologists and neurosurgeons at large. Bonica (1991) found that, of 26,281
pages of text in 14 textbooks of neurology, neurosurgery, medicine and surgery, only
6.5 (0.025%) dealt specifically with CP, a situation that persists almost unchanged to
this day. Consequently, most physicians in practice have little or no awareness of the
subject.
Until the mid 1980s, little or no research on the clinical characteristics as well as
the basic mechanisms and pathophysiology of CP was done, with only a handful
of basic and clinical scientists devoting efforts to these objectives. Not even the
establishment of the IASP in 1973 and of the journal Pain in 1975 changed this
dismaying panorama. At the end of the 1990 Ann Arbor symposium on central pain
syndromes (Casey 1991), Lindblom epitomized the problem: The pain mechanisms
of central pain syndromes are virtually unknown and specific analgesic measures
are lacking for the vast majority of patients: CP remained a puzzling mystery
(Pagni 1989).
The extent of the puzzle is given by the bewildering array of theories proposed
over 100 years, several directly contradicting one another:
1) Irritation of cells and fibers of spinothalamic and lemniscal systems.
2) Irritation of the sympathetic system, outside the CNS, central cerebrospinal
sensory pathways being destroyed.
3) Diversion of pain impulses to the hypothalamus.
4) Summation and wrong integration of pain impulses on a few spared nociceptive neurons.
5) Loss of inhibitory pain mechanisms exerted by thalamus, cerebral cortex,
striopallidum, medial lemniscus, brainstem.

Central Pain Syndrome

6) Activation of alternative secondary pathways, not usually opened and not used
when conduction via the spinothalamic complex is available.
7) Abnormal spontaneous or provoked activity in deafferented central sensory
neuronal pools which may act as spontaneous dysesthesia and pain-generating
mechanisms.
8) Hypersensitivity of deafferented medial midbrain tegmentum, posterior
thalamus, thalamic radiations and somatosensory cortex.
9) Activation of nonspecific polysynaptic pathways (paleospinothalamic system),
i.e., the neospinothalamic complex and lemniscal system being damaged,
nociceptive stimuli are conveyed to the conscious level on this diffuse network
of short neurons.
Much has changed over the past 15 years, with several groups applying modern
neuroimaging and neurophysiologic techniques to the study of CP. In particular, it is
our contention that an explanation and a cure for this enigma can now be offered.

CENTRAL PAIN OF BRAIN ORIGIN

EPIDEMIOLOGY AND CLINICAL FEATURES

1. Lesions causing CP and location (Table 2.1a,b)
BCP has been caused by all kinds of lesions at any level along the spinothalamoparietal path, from brainstem to cortex, a fact already appreciated in the 1930s
(Garcin 1937; DeAjuraguerra 1937; Riddoch 1938). These include rapidly or slowly
developing processes, apparently without differences in probability of triggering
CP (but systematic studies have not been conducted), compressive or disruptive/
distractive (these latter perhaps being more often associated with CP).
Stroke, either hemorrhagic or ischemic, is the commonest cause of BCP (without
differences between the two); dismayingly, iatrogenic CP is not rare. In agreement
with their known incidence, in all studies, infarcts are more common than
hemorrhages (roughly 4:1).
When the lesion is thalamic, Vc is always involved (the case of Gonzales et al.
[1992] had signs of capsular involvement). Contrary to previous belief, one third or
even less of BCP cases are purely thalamic (e.g., Hirato et al. 1993; Andersen et al.
1995; Tasker 2001b; Widar et al. 2002; Oliveira et al. 2002; see also Schmahmann
2003) and complete thalamic syndromes are exceptional. CP does not arise following
thalamic lesions only damaging the kinesthetic afferent pathway and probably the
spindle afferent pathway as well (Ohye 1998). In all other cases, lesions are corticosubcortical, in the brainstem, capsulothalamic or lenticulocapsular, or diffuse. Most
CPSP is supratentorial (roughly 80%; Tasker 2001).
All cortical lesions responsible for CP involve, exclusively or in combination, the
parietal lobe, and specifically SI (and also SII) (e.g., Bassetti et al. 1993). Pain occurring
acutely immediately after traumatic cortical injury (e.g., penetrating head injuries)
a
lancinating pain felt by the patient at the very moment of injury
has been considered
CP of cortical origin (Garcin 1937): it fades away rather quickly (hours to days).
The most common site of brainstem lesions (either stroke or hematobulbia,
syringobulbia, tumors and MS) is the medulla oblongata, with few cases of pontine
and no pure midbrain spontaneous CP having been reported. However, this may
actually be an underestimation, as a brainstem lesion was found in 70% of stroke
patients in whom MRI was performed (Vestergaard et al. 1995). CP of bulbar origin
is generally due to thrombosis of the posteroinferior cerebellar artery (PICA) giving
rise to Wallenbergs syndrome, in which a lesion impinges on the spinothalamic
9

10

Central Pain Syndrome

Central Pain of Brain Origin: Epidemiology and Clinical Features

11

12

Central Pain Syndrome

Central Pain of Brain Origin: Epidemiology and Clinical Features

13

14

Central Pain Syndrome

Central Pain of Brain Origin: Epidemiology and Clinical Features

15

16

Central Pain Syndrome

Central Pain of Brain Origin: Epidemiology and Clinical Features

17

18

Central Pain Syndrome

Central Pain of Brain Origin: Epidemiology and Clinical Features

19

20

Central Pain Syndrome

Central Pain of Brain Origin: Epidemiology and Clinical Features

21

22

Central Pain Syndrome

tract and on the nucleus and/or the descending root of the trigeminal nerve on the
same side.
2. Incidence and prevalence (Table 2.2)
For more than a century, BCP has been considered rare, based on sheer opinion (1
out of 1500 strokes according to Davis and Stokes 1966), a concept upheld by the
rapid retrospective survey on 400 stroke patients (with unclear definition of source
population under consideration) performed by Bowsher (1993), 2% of whom (out
of 25% with somatosensory deficits) developed CP. Several studies dealing with
this question have been published; unfortunately, the vast majority has a strong
selection bias, being retrospective in nature or drawing from hospitalized patients in
single neurology departments (i.e., not mirroring the true prevalence in the general
population). One confounding factor has also been the delayed onset in several
patients (up to years; Table 2.5), leading to underestimations.
Luckily, prospective studies conducted over the past decade overturned this
misconception. Today, we know that no less than 8% of all strokes (brain and
brainstem, ischemic and hemorrhagic) originate CP/CD/CNP; this figure rises to
roughly 20% if somatosensory signs are present.
Almost 5 million Americans (2.4% of adults) had a stroke, with almost 600 000
new survivors each year. Stroke prevalence in the EU is about 1100/100 000. An
estimated 15 million people worldwide survive a minor stroke each year (WHO
estimate 2002). As stroke attacks seems to be lower in Western than in Eastern
Europe or China for both men and women and blacks have a higher incidence of

Central Pain of Brain Origin: Epidemiology and Clinical Features

23

24

Central Pain Syndrome

Central Pain of Brain Origin: Epidemiology and Clinical Features

25

26

Central Pain Syndrome

Central Pain of Brain Origin: Epidemiology and Clinical Features

stroke than whites (at least in the United States and the UK), the worldwide
prevalence of CPSP alone may amount to several million patients.
As regards other causes of brain lesions, trauma, particularly penetrating head
injury, is said to be rarely associated with CP (2 of 11 patients with somatosensory
abnormalities in a 1000 patient series; Marshall 1951). No prospective study exists,
although clinical experience would suggest that CP in such cases is indeed rare. CP
arises only after disruptive or compressive lesions along the spinothalamoparietal
axis; contusions mainly involve basal frontal and temporal areas and hematomas

27

28

Central Pain Syndrome

(extradural and subdural) rarely impinge on the parietal lobe to such extent to
disrupt it. Moreover, half of patients with severe head injury go on to die within days
or weeks of trauma or develop severe disability. And yet a recent Chinese series
(Li 2000) reported 20 cases of typical CP after severe head injury observed over
8 years. At CT, 16 had brain hematomas at cortical and subcortical levels (14 were
evacuated) and 4 brain infarctions in deep nuclei and corona radiata. It is quite
possible that the true incidence is actually underestimated.
In the United States, 2 million cases severe enough to cause brain damage occur
yearly (20% of all head injuries), but penetrating trauma has become rare after the
enforcement of helmets and the introduction of airbags. Even in a war setting,
helmets reduce the extent of damage and the limited extent of SI (the portion of the
parietal cortex that has to be involved) ensures that CP would be rare.
Cancer most certainly represents an important cause of CP and, although data are
very limited and in the absence of prospective studies, we may expect that roughly
2% of cancer patients suffer CP. Parietal tumors not infrequently trigger CP
(especially meningiomas, but also gliomas, rather than metastases) (e.g., Bender and
Jaffe 1958), but this goes often unrecognized. Thalamic tumors only rarely cause CP
(Tovie et al. 1961), but this is not surprising: striatal tumors also only rarely cause
extrapyramidal symptoms (see also Lozano et al. 1992). Worldwide 10 million cases
of new cancers occur every year.
An underrecognized cause of CP is surgery (and particularly neurosurgery;
Table 2.13), either via direct brain damage or strokes.
No quantitative data exist regarding differential incidences of CP between rapidly
and slowly developing lesions.
In sum, in the United States alone, there should be more than a half million people
suffering BCP.
3. Age of onset of CPSP (Table 2.3)
Although one prospective study on CPSP found no significant difference between the
general stroke population and CPSP, the general impression from all other studies,
both prospective and retrospective, is that CPSP affects younger patients (sixth
decade versus seventh decade). However, this awaits confirmation.
4. Sex distribution of CPSP (Table 2.4)
Stroke is more common in men than in women (but deadlier in the latter)
worldwide. We should thus expect more men than women complaining of CPSP.
A prospective study showed that, although men were more affected than women by
stroke, women were more affected than men by CPSP; however, the difference was
not statistically significant. Other studies are at odds between them, but Nasreddine
and Savers systematic review on thalamic CPSP found a larger percentage of men
suffering CPSP. Wallenbergs strokes are also more frequent in men, and CPSP also
appear to predominate in them.
5. Time to pain onset (Table 2.5)
CPSP and CP in general can develop immediately or up to 10 years after the inciting
event. In several cases, CPSP is a presenting symptom (roughly one fourth): we had a

29

CVA 59.2 (40
78, median 59)

SAH 46.1 (29
62, median 45)

104 (thalamic
hematoma)
175
CVA 111
SAH 19

Mori et al. (1995)

Chung et al. (1996)

Bowsher (1996)

(57.5 (47
76, median 55.5))

(58.8 (48
70, median 56))

57.8 (35
86, median 56)

60.4 (42
75, median 59.5)

25
41 LMI
14 MMI

McGowan et al. (1997)

Paciaroni and Bogousslavsky (1998)

Kim and Choi-Kwon (1999)

61.4 (39
89, median 63)

58 (24
84)

(continued)

61.2 (39
80, median 63)

60 (40
78)

60 (median) (18
85)

63

Nasreddine and Saver (1997)

67 and 54 (2 pts)

17 (brainstem
stroke)
180

61 (range 52
71)

67

Kim and Choi-Kwon (1996)

nr

Kim and Bae (1997)

60 (27
83)

61.2 (11
93)

68.7 (putaminothalamic
hemorrhages)
63.8 + 10.7

71.5 (median) (41
78)

139

69 (46
86) PSS

69 (56
85, median 67)

46.3 (28
58, median 48)

Lampl et al. (1995)

69 (median) (40
79)

67

66 (50
86, median 64.5)

39

18

Wessel et al. (1994)

87

6 (parietal CP)

Schmahmann and Leifer (1992)

450 (67.2% of pts)

59.2 (39
75, median 51)

62.6 (52
73, median 63)

55.7 (19
78, median 63)

CPSP group

Samuelsson et al. (1994)

73

Tasker et al. (1991)

56.8 (20
86, median 63)

63.2 (19
81, median 66)

Stroke group

Age of pts (years, mean)

Andersen et al (1995)

40
12 (parietal CP)

Bogousslavsky et al. (1988)

25

Graff-Radford et al. (1985)

Michel et al. (1990)

Number of pts

Author

TABLE 2.3. Age of onset (CPSP only)

30
13
157

107
43

Kong et al. (2004)

Widar et al. (2004)

18 thalamocortical

Kameda et al. (2004)

10 brainstem

Greenspan et al. (2004)

54.7 (26
72, median 55); nr in

2 pts Spinal vs. thalamocortical
difference statistically
significant (ANOVA, p 0.019)

12 spinal

Garcia-Larrea et al. (2002)

43

49.5 (36
67, median 42.5)

119

Weimar et al. (2002)

Widar et al. (2002, 2004)

44.2 (25
61, median 47.5)

39

MMI 65.0 + 12.3 (difference statistically

significant, p < 0.034; no significant
difference in the age of onset between
men and women)
60.9 yrs

Medullary infarct, overall:

61.3 + 12.4 yrs
LMI 60.7 + 12.4 yrs

66 (33
82)

64.5 yrs (median,

overall 63, range: 14
91)

73.5 (40
96, median 74)

Bowsher (2001)

Lampl et al. (2002)

13 women: 76 (range 54
82)

30 men: 64 yrs (range 33
79)

67.6 (range 33
82)

62.5

nr

60.9 (43
78, median 64)

62 (median)

60.5

77 (69
91, median 77.5)

54

59.7 (35
85, median 58)

16

CPSP group

Kumral et al. (2001)

Stroke group

Number of pts

Age of pts (years, mean)

Author

TABLE 2.3 (continued)

31

Pathology

Michel et al.
(1990)
Tasker et al.
(1991)

Kameyama
(1976)
Graff-Radford
et al. (1985)
Kawahara
et al. (1986)
Bogousslavsky
et al. (1988)
Mauguie`re and
Desmet (1988)
Leijon et al.
(1989) and
Boivie et al.
(1989)

2 (67%)

4 (33.3%)
33 (45%)

8 (66.7%)
40 (55%)

7 (26%)

21 (52.5%)

20 (74%)

1 (33%)

1 (25%)

4 (36%)

CPSP

Stroke (same goup of CPSP pts for

2 different studies. Authors statement:
male dominance could not be explained by
a difference in the incidence of stroke
between men and women because stroke
affected more women than men in Sweden.
However, strong bias likely, given the
selection criteria
Cortical pain (MRI or CT confirmed lesion).
12 pts
CPSP. 73 personal consecutive cases

19 (47.5%)

Thalamic infarcts

21 (56.8%)

11 (44%)

61 (70%)

Stroke

9 (37%)

16 (43.2%)

Small (<2 cm) thalamic hemorrhages

3 (75%)

7 (64%)

CPSP

Women

15 (63%)

14 (56%)

Non-hemorrhagic thalamic infarctions

Men

Thalamic vascular lesion

26 (30%)

Stroke

Thalamic vascular lesions

Head and Holmes Thalamic syndrome/LMI

(1911)

Author

TABLE 2.4. Sex distribution (CPSP only)

0.9

0.76

1.27

Stroke

(continued)

1.21

2.8

1.66

0.5

0.42

1.75

CPSP

Men/women ratio

32

Mori et al.
(1995)

Lampl et al.
(1995)

Kumral et al.
(1995)
Andersen et al.
(1995)

Wessel et al.
(1994)
Samuelsson
et al. (1994)

Parietal lesion

Schmahmann
and Leifer
(1992)
Steinke et al.
(1992)

139 putaminothalamic
hemorrhages (among 279
consecutive pts with supratentorial
hemorrhages)
104 thalamic hemorrhages

110 (53%)

Acute stroke (prospective study)

63 (60.5%)

Pts with
44 (50.5%)
somatosensory
deficits (SSD)
88 (63.3%)

34 (34%)
7 (44%)

5 (PSS)

20

3 (50%)

CPSP

nr

27 (61%)

Hemorrhage

Men

6 (33%)

6 (33%)

Infarct

33 (53%)

Stroke

Thalamic stroke (CP in 10/18 pts,

sex distribution not reported)
Lacunar infarct syndrome. Pure
sensory stroke (thalamic) in
10 cases. CSPS in 3 pts
Thalamic hemorrhages

Thalamic stroke

Pathology

Author

TABLE 2.4 (continued)

41 (39.5%)

1.5

1.7

51 (36.7%)

1.13

0.51

1.05

0.5

1.58

0.5

1.13

Stroke

1.02

9 (56%)

5 (PSS)

nr

3 (50%)

CPSP

0.77

1 (PSS)

CPSP

Men/women ratio

43 (49.5%)

97 (47%)

66 (66%)

19

12 (66%)

17 (39%)

12 (67%)

29 (47%)

Stroke

Women

33

75 (43%)
13 (76.4%)

Thalamic hemorrhages

Pure or predominant sensory stroke due to

brainstem lesion. 17 pts

25 (64%)
3 (75%)

Brain CP (MCS pts series)

Pure sensory thalamic stroke

10 (62.5%)
64 (67.3%)

Bilateral thalamic infarction

Stroke

5 (62.5%)

1 (6.2%)

31 (32.6%)

6 (36.5%)

LMI 10
12 (21.8%)
(71.4%)
MMI 3 (60%)

Severe post-stroke paresthesias in lateral

and medial medullary infarction

43 (78.2%)

40 (55%)

8 (32%)

18 (29%)

4 (23.6%)

Subgroup of Bowsher (1996) pts

17 (68%)

63 (55%)

Meta-analysis

9 (56%)

Nasreddine and
Saver (1997)
Yamamoto et al.
(1997)
Paciaroni and
Bogousslasky
(1998)
Bowsher et al.
(1998)
Kim and
Choi-Kwon
(1999)

45 (71%)

LMI (retrospective analysis)

2 (100%)

McGowan et al.
(1997)

Kumral et al.
(2001)
Bowsher
(2001)

100 (57%)

0.75

2.12

2.5

3.25

3 (37.5%)

2.06

1.66

LMI 4
3.58
(28.6%)
MMI 2 (40%)

33 (45%)

1 (25%)

14 (36%)

51 (45%)

7 (44%)

(continued)

1.6

1.5

2.5

1.21

1.78

1.23

1.28

3 (100%)

Chung et al.
(1996)
Kim and Bae
(1997)

0.53

11 (48%)

12 (52%)

SAH/postneurosurgery
Brain trauma

1.33

1.19

48 (43%)

63 (45%)

63 (57%)

75 (54%)
CVA

Personal series of 138/156 CP pts

Bowsher (1996)

34
nr

CPSP

Lateral and medial medullary infarct

Greenspan et al.
(2004)
Kameda et al.
(2004)

Kong et al.
(2004)
Widar et al.
(2004)

10 (77%)

Stroke

Widar et al.
(2002, 2004)

68 (64%)
30 (70%)

Stroke

nr

nr

13 (30%)

39 (36%)

Overall: 55 (26%)

Overall: 159 (74%)

Stroke

LMI MMI: 1 (17%)

LMI MMI: 5 (83%)

LMI: 45 (27%)

13 (30.2%)

42 (35%)

MMI: 9 (22%)

8 (73%)

Stroke

Women

MMI: 32 (78%)

LMI: 122 (73%)

30 (69.7%)

77 (65%)

Stroke

Weimar et al.
(2002)

16 (41%)

Thalamic ventroposterior stroke

CPSP

Lampl et al.
(2002)

Stroke

Pathology

Men

Author

TABLE 2.4 (continued)

nr

nr

nr

3 (23%)

3 (27%)

23 (59%)

CPSP

2.3

1.74

Overall: 2.89

LMIMMI: 5

MMI: 3.5

LMI: 2.7

1.83

Stroke

nr

nr

nr

3.3

2.66

0.69

CPSP

Men/women ratio

35

1 pt

This author rejected, on the basis of published reports, the Lhermitte (1925) hypothesis
that pain is apt to appear earlier with hemorrhages than with ischemic lesions
In pontobulbar strokes pain usually coincides with stroke, but in some cases it may
appear several days or more after stroke
Cortical lesion

(continued)

Case report and review. Cortical pain, 8 pts, 2 missed data. Time span from immediate
to 6
7 mos

Fields and Adams (1974)

100%

One pt developed central pain immediately after stereotactic coagulation at thalamic

level

Cassinari and Pagni (1969)

Reversible central pain (cortical origin)

1 pt

(100%?)

Pain can arise during the acute period of the stroke, at the same time of the
development of the hemiplegia, or can be delayed, arising some months after the
stroke
Pain was the first symptom of stroke and preceded motor deficits in cases 5, 7 and 8; it
developed early (hours) after the thalamic stroke in case 9, but it was delayed (days,
weeks or months) in cases 10, 11, 12 and 13
CP in pontobulbar lesions is generally delayed (months) after the stroke (cases 2 and
3)
Wallenbergs syndrome (LMI): pain can be an initial feature, but in many cases
develops days or months after the stroke, arising at the same time in face and
contralateral limbs; often limb pain shows a delayed (weeks) onset compared to face
pain
Thalamic pain: it can be immediate or delayed more than one year after stroke

2 wks

Cerebral AVM. CP appearance after 8 yrs

50%

87%

1 pt

2 days
1 mo 1 mo
3 mos 43 mos Notes

Hamby (1961)

17%

50%

1 day

Silver (1957)

Biemond (1956)

Riddoch (1938) (review)

Garcin (1937) (review)

Alajuanine et al. (1935)

Head and Holmes (1911)

Dejerine and Roussy (1906)

Edinger (1891)

Author

Cumulative % of pts with CP within

TABLE 2.5. Timespan between stroke injury and CP onset

36

15%

50%

29%

(17%)

Leijon et al. (1989)

Michel et al. (1990)

Tasker et al. (1991)

Schmahmann and Leifer

(1992)

100%

Kumral et al. (1995)

0%

100%

Samuelsson et al. (1994)

50%

33.3%

46%

58.3%

52%

33%

75%

96%

91.6%

78%

100%

70%

0%

Bogousslavsky et al. (1988)

0%

50%

Mauguie`re and Desmet (1988)

Wessel et al. (1994)

63%

Shieff and Nashold (1987)

75%

75%

Schott et al. (1986)

100%

100%

100%

100%

100%

100%

Series of 73 pts with CP of brain origin. Pain onset was delayed after an identifiable
ictus in about two thirds of the cases, being less than 1 yr in 51% of the pts. The onset
was immediate in 29% but 42 yrs in 4.2% of their cases
Parietal pseudothalamic syndrome. 6 pts. Pain occurrence: 1 wk after a traumatic
injury, 4 mos after resection of meningioma, and 3 wks, 6 mos, 1 and 4 (or 5) yrs after
vascular accidents. Immediate hemiparesthesia in 1 case
CPSP pts. Median time span: 3 mos. In some cases pain developed 24 mos after the
stroke
10/18 pts with single thalamic infarct and CP. Pain onset within 1 mo in 5 pts; within 6
mos in all pts
3 CPSP pts among 39 pts with lacunar infarct syndromes (pure sensory stroke).
Painful dysesthesia developed gradually during the 1st mo after stroke
Prospective study. 100 pts with thalamic hemorrhage. Delayed (1 mo) thalamic pain:
3 pts
Delayed (1 mo) thalamic pain plus chorea plus ataxia (thalamic syndrome): 6 pts

Series of 27 pts. Pain onset: 4/27 on 1st day; 14/27 within 1 mo; 21/27 within 3
mos; 24/27 within 1 yr; 27/27 within 34 mos
Retrospective study. 12 pts. Cortical CP. Lesion sparing the thalamus confirmed by
means of MRI or CT scans with reconstruction. Very delayed onset (2 yrs) in 1 pt

Delayed onset (1 wk, 2 mos and 3 mos) of CP in 3/40 pts

Thalamic stroke. Immediate pain development in 15/20 pts. Pain onset delayed until
10 yrs after the stroke in one case. Data in text and in tables not in agreement
Series of 27 pts with the thalamic syndrome. 17/27 pts had their pain from the time of
their initial stroke, while in 10 it developed after an interval of as long as 2 yrs, in 3
after a further stroke
Thalamic syndrome, 30 pts. CP in 24/30 pts. Max. delay: 6 mos

2 days
1 mo 1 mo
3 mos 43 mos Notes

Cumulative % of pts with CP within

Bowsher (1993)

1 day

Author

TABLE 2.5 (continued)

37

18%

Chung et al. (1996)

Nasreddine and Saver (1997)

Bowsher (1996)

Kumral and Celebisoy (1996)

Andersen et al. (1995)

56%

(see notes)

40% (<1 wk)

63%

71%

100%

100%

100%

Systematic review on thalamic syndrome. 180 pts with the

documented by postmortem examination. Exclusion criteria:
nonvascular etiology, and symptoms consisting solely of evoked
spontaneous pain. Information regarding the time from stroke to
pain available in 66 cases.
Pain onset (% of pts, noncumulative):
immediate: 18%
within the first post-stroke wk: 18%
1 wk to 1 mo: 20%
1
3 mos: 15%
3
6 mos: 12%
6
12 mos: 6%
41 yr: 11%

dorsal thalamic lesion: 25% of pts

posterolateral thalamic lesion: 32% of pts

(continued)

site of the lesion

pts with tumoral/
dysesthesia without
the development of

Thalamic syndrome present at discharge from hospital (14 days after the admission),
on average:
posteromedial thalamic lesion: 25% of pts

Thalamic hemorrhage. Paresthesia at onset: posterolateral lesion: 31/77 pts (40%).

(Disagreement between data in text and in Fig. 8 [Fig. 8: 21% of paresthesia at onset])
dorsal lesion: 34% of pts

SAH pts (16 pts): 4.4 mos (range 0.25
12, median 4)

CVA pts (63): 6.2 mos (range 0.08
72, median 3)

Personal series of 138/156 CP pts. Pain onset immediate (arguably less than a week
from the inciting event) in 47 (8 N/K, abbreviation not explained in the paper) pts
(39.8% of cases). Delayed pain onset (from 1
2 wks up to 6 yrs) in 83 pts (60.1%)
Average interval between stroke and delayed pain onset:

Prospective study. The number of pts suffering from CP increased in time from 10 at 1
mo to 13 at 6 mos and to 16 at 1 yr after the stroke (end of follow-up)
1 thalamic stroke. CPSP 8 yrs after stroke

38

Oliveira et al. (2002)

Bowsher (2001)

Kumral et al. (2001)

Li (2000)

Kim and Choi-Kwon (1999)

0%

100% MMI

14.3% LMI

75%

0%

42.9% LMI

75%

100%

100%

85.8% LMI

80%

Kim (1998)

100%

75%

100%

100%

100%

100%

100%

Gradual onset in 77.5% of pts

1 pt with possible CPSP was eliminated from study because pain appeared 108 mos
(9 yrs, 4 yrs in the text) after stroke

16 pts with bilateral thalamic infarction. CPSP in 1 pt 3 mos after the stroke

LMI pts: onset of sensory symptoms immediate to 6 mos after the stroke. Among 14
pts reporting severe paresthesia/pain the onset of symptoms was immediate in 2
cases (14.3%), within 1 mo in 4 cases (28.6%) and between 1 and 3 mos in 6 cases
(42.9%). In one pt (7.1%) symptoms developed after 6 mos and in one case the time
span between stroke and the onset of symptoms was unknown. MMI pts: immediate
development of sensory symptoms, if present, in all but 1 pt in whom they developed
within 1 mo
3
120 days after severe head injury (mean: 30.5 days)

6 stroke pts. Early-onset CPSP contralateral to the ischemic lesion delayed ipsilateral
sensory symptoms mirroring the site of the most severe CPSP (without new ischemic
lesions). In 1 pt unclear time span
Medullary infarction

Lateral medullary infarction (LMI). CPSP in 16/63 pts. Average time span between LMI
and pain onset: 4 weeks (range 1
24). Pain onset was within 2 weeks in 3 pts, 1 mo
in 9 pts and 6 mos in the remaining 4 pts
39 CPSP pts. Interval between the onset of the inciting lesion and the occurrence of
pain: 1
6 mos
25 pts with pure sensory syndrome in thalamic stroke
CPSP in 4/25 pts after a mean of 10.5 days (range 2
15 days). Delayed pain in 2 pts,
delayed pain and dysesthesias in 1 pt and delayed dysesthesia in 1 pt
9 LMI CPSP. Median stroke-pain delay 4 mos (range: 3
41)

2 days
1 mo 1 mo
3 mos 43 mos Notes

33%

40%

1 day

Cumulative % of pts with CP within

Peyron et al. (1998)

Paciaroni and Bogousslasky

(1998)

Yamamoto et al. (1997)

McGowan et al. (1997)

Author

TABLE 2.5 (continued)

39

53%
8 pts

Hansson (2004)

Nakazato et al. (2004)

14%

8%

Greenspan et al. (2004)

Rossetti et al. (2003)

Kim (2003)

Lampl et al. (2002)

46%

43%

8 pts

100%

100%

100%

14 pts. Wallenberg syndrome, unreliable data (8 8 pts instead of 14)

Stroke pts. Data from Widar et al. (2002), Andersen et al. (1995) and Leijon et al.
(1989)

20 pts with CPSP or paresthesia after lenticulocapsular hemorrhage. Pain or

paresthesia occurred 0
24 mos after the onset, more prominently in the leg than
other body parts
1 pt. Ipsilateral acute limb pain at stroke onset. MRI: infarction in the territory of the
right anterior parietal artery. Basal ganglia, thalamus, and subthalamic region intact.
Incidence of acute ipsilateral pain at the onset of stroke: 1/29 pts with hemiballismushemichorea (retrospective analysis of more than 4000 pts). Occurrence of
hemiballismus with ipsilateral pain is specific for an anterior parietal artery stroke
13 consecutive CPSP pts. Mean latency from stroke to CPSP: 5.6 mos (1
12, median
4)

Prospective study. Primary end point of the study: evaluation of CPSP incidence and
time span between stroke and pain. CPSP in 7/39 pts
Day CPSP occurred: 12, 36, 67, 70, 209, 246, 267. Median: 70

40

Central Pain Syndrome

patient who immediately suffered CP upon awakening from parietal glioma surgery;
in one patient described in Garcin (1968), the onset was so sudden that the patient
thought he had been hit on the face and Cassinari and Pagni (1969) reported one
patient who developed CP immediately after stereotactic coagulation at thalamic
level. However, a majority (half to three quarters) develops it within 3
6 months
after the causative lesion. Pain onset delayed over 1 year is rare, but not exceptional:
in such cases, the pain may sometimes commence after an infection, trivial accident
or surgery (Tasker and Dostrovsky 1989). In some patients, the onset coincides with
improvement of the sensory loss. The time of onset does not appear to depend on
lesion level and early-onset (including immediate) and late-onset pains appear to be
clinically identical. CP may also precede other neurological signs.
6. Side of the lesions (Table 2.6)
Right-sided lesions predominate among CPSP patients at both thalamic and cortical
levels. This difference is most likely not due to a difficulty of communication after left
lesions (moreover, right lesions may cause hemineglect and anosognosia more
frequently). It should not depend on simple prevalence of right strokes either, since
men, but much less women, show CPSP laterality.
7. Size of the lesion and CPSP (Table 2.7)
Data are available only for thalamic vascular strokes. The hypothesis that CP
correlates with the size, rather than the site, of the lesion is a time-honored one, but
available data are conflicting. There are several old reports in which the size of the
thalamic lesion originating CP was noted at autopsy. Lhermitte (1936) suggested that
CPSP is rare in patients in whom the thalamus is completely or almost completely
destroyed by a large hemorrhagic lesion. However, Garcin (1937), quoting Lhermitte
and Schuster, stated that the volume of the lesion did not seem to correlate with the
presence or the absence of CP in thalamic syndrome. Nevertheless, he noted that CP
develops more frequently when the thalamic nuclei are affected by larger or multiple
hemorrhages, especially with involvement of the lateral nucleus. More recently, after
the introduction of new imaging technologies (CT, MRI), the site and the volume
of the lesions have been reported in papers dealing with thalamic stroke. In some
papers the authors also reported the occurrence of CPSP, allowing the evaluation of
the lesion volume in CPSP cases. Apparently, the volume of the lesion in patients
with thalamic CPSP does not seem to differ from the expected volume in thalamic
hemorrhage, nor between patients with somatosensory deficits with and without
CPSP. In conclusion, it would seem that CPSP can follow both small and large lesions
and the site of the lesion is more important than its size. However, other data
strongly suggest that total destruction of the thalamus is incompatible with a CP
generator on that side (Chapter 7).
8. Pain distribution
Contrary to the notion that CP is diffuse and difficult to localize, patients can usually
describe the location of their pain. Its distribution corresponds somatotopically to
the site of the lesion: e.g. after lower medullary infarction (Wallenbergs syndrome),
CP, when present, is projected to the ipsilateral hemiface, tongue, gums and inner

Central Pain of Brain Origin: Epidemiology and Clinical Features

41

42

Central Pain Syndrome

Central Pain of Brain Origin: Epidemiology and Clinical Features

43

44

Central Pain Syndrome

45

Mean volume of the lesion

nr

4 mL or less

32.7 mL (range 0.5
144.4)

BS: 2.6 mL (range 2
4.3)
TH: 18.1 mL (range 0.6
144.4)
ET: 73.2 mL (range 47.2
99.2)
UN: 0.8 mL (range 0.5
3.

33% < 4 mL
67% 4 4 mL

Lacunar infarcts

Median volume of the lesions:

pts without SSD: 2 mL (range 0
96)
Pts with SSD without pain: 1 mL
(range 0
200)
CPSP pts: 16 mL (range 0
210)

Author

Kameyama (1976)

Kawahara (1986)

Leijon et al. (1989)

Kumral et al. (1995)

Samuelsson et al. (1994)

Andersen et al. (1995)

TABLE 2.7. Size of the lesion

(continued)

Authors conclusion: Pts with CPSP have larger lesions since larger lesions affect by
chance the spino-thalamo-cortical pathway more often.

Prospective study
Volume of the lesion evaluated on CT scan. Larger lesions in CPSP pts (difference
statistically significant), but no demonstrable difference in single acute lesion size
between CPSP pts and pts with sensory deficits (SSD) but without pain

39 pts. 3 CPSP in 10 pts with pure sensory stroke (posterolateral thalamus). Mean
diameter of the lesions:
Pure sensory stroke (9 lesions): 9 mm
Sensorimotor stroke (9 lesions): 15 mm
Pure motor stroke (14 lesions): 16 mm

Thalamic hemorrhage. 100 pts. Small hemorrhages in 33 pts; large hemorrhages

(42 cm in diameter and/or 4 mL in volume) in 67
CPSP in 9 pts (8 with large hemorrhages and 1 with small hemorrhage)
6 posterolateral hemorrhages (1 small, 5 large) in 6 pts, anterolateral, medial
(small) and dorsal (large) in 3 pts

Group of 27 CPSP pts. BS: brainstem lesions (8 pts). TH: thalamic lesions (9 pts).
ET: supratentorial extrathalamic lesions (6 pts). UN: unidentified supratentorial
lesions (4 pts)
Thalamic lesions extended beyond the thalamus in 7/9 cases

6 cases of paresthesia and/or dysesthesia among 37 pts with small (less than 2
cm in mean diameter) thalamic hemorrhages. Assuming approximately spherical
lesions, their (calculated) volume was about 4 mL or less

In his series of 11 pts with VPL lesions extending to nucleus centrum medianum
thalamic pain never developed
Authors conclusion: CP is rare in cases with large thalamic lesions.

Notes

46

Volume of hematoma (mm ) among

survivors:
Within thalamus: 3 + 3
Extending to IC: 9 + 6
Extending to midbrain or putamen: 18 + 12

Mean length of the hematomas largest

diameter and corresponding volume:
anterior type: 17 mm
(95%CI 10.6
22.5); 2.5 mL
posteromedial type: 25 mm
(95%CI 21.4
28.1); 8.2 mL
posterolateral type: 31 mm
(95%CI 28.4
32.9); 15.5 mL
dorsal type: 24 mm
(95%CI 21
25.9); 7.2 mL
global type: 37 mm
(95%CI 24.2
40.9); 26.4 mL

nr

nr

Chung et al. (1996)

McGowan et al. (1997)

Paciaroni and Bogousslasky (1998)

Mean volume of the lesion

Mori et al. (1995)

Author

TABLE 2.7 (continued)

Pure sensory thalamic stroke. 25 pts

CPSP in 4/25 pts
Thalamic infarcts were very small. Pts without CPSP had a median involvement of 2
nuclei (range 1
5)

LMI 52 pts. Extent of the infarction (MRI scans) evaluated in a blind fashion. Size of
the infarction graded 0
4 according to a predetermined scoring scale (0 normal,
4 LMI plus ventrobasal cerebellar infarction plus brainstem infarction beyond the
lateral medulla). Statistical analysis performed
Authors conclusion: The different MRI infarct size scores had no relationship with
CPSP.

Thalamic hemorrhage. 175 consecutive pts. Mean volume calculated assuming a

spherical shape
Anterior type: 11 pts; posteromedial type: 24 pts; posterolateral type: 77 pts; dorsal
type: (32 pts; global type: 31 pts. The global type was large enough to occupy
nearly the entire area of the thalamus
Paresthesia at onset:
posterolateral type: 21% (as reported in fig. 8b, in text it is stated 31/77 pts, i.e.
40%)
dorsal type: 34%
Delayed thalamic syndrome:
posteromedial type (volume 8.2 mL): 25% of pts
posterolateral type (volume 15.5 mL): 32% of pts
dorsal type (volume 7.2 mL): 25% of pts

Statistically significant difference among all hematoma size

Thalamic ematoma. 104 pts. 86 survivors at 6 mos. CPSP in 6 pts (within thalamus
hematoma: 2, extending to the IC: 3, extending to midbrain and putamen: 1)

Notes

Central Pain of Brain Origin: Epidemiology and Clinical Features

aspect of cheek and often, but not always, contralaterally below the collarbone,
diffusely or sectorially (arm and rarely the trunk and leg), simultaneously or at
different intervals, in keeping with damage to the spinal nucleus and tract of the
trigeminal nerve and the crossed STT. Exceptional simultaneous involvement of both
sides of the face is explained by involvement of the descending root of the trigeminal
nerve of one side and the crossed quintothalamic pain fibers coming from the other
side (Riddoch 1938).
CP is always segmentally distributed, supporting a role of somatotopically
organized structures. Roughly 40% of all BCP patients complain of hemibody pain
(hemipain), with or without the hemiface. In all other cases, CP is restricted to one or
more body parts, e.g., the hemiface, one hand, one foot, a quadrant of the body, or
the mouth and hand (the cheiro-oral syndrome), without a transition zone; the face
and arm are most affected, and the leg least, reflecting greater cortical representation,
but a hemiface singly is affected in roughly 10% of the cases. The pain may vary in
site (wander), disappearing from one limb only to arise in another, and intense
pains in the limbs may be found simultaneously with only paresthesias in the face,
or vice versa (Garcin 1937; Riddoch 1938).
The area of pain (spontaneous and evoked and rarely only evoked) may match the
sensory and/or motor deficit, but may also be patchy, i.e., confined to a fraction of
the disabled region, even after lesions causing extensive loss of somatic sensibility
(e.g., Michel et al. 1990; Tasker et al. 1991); in contrast, CP is never localized to an
unaffected area. CP is experienced as superficial (projected to the skin), deep
(originating in muscle and bone) or both in varying proportions.
9. Quality of pain (Table 2.8)
Most patients experience one or more pain qualities simultaneously (two to four), in
the same or different body regions (e.g., burning in leg and aching in face or, for
example, in Wallenbergs syndrome dysesthesias to the hemiface and shooting pains
to the limbs and trunk or vice versa) and seemingly identical lesions may cause
different combinations of pain qualities in different patients. CP can have any quality,
although some qualities are commoner; bizarre qualities are the exception rather than
the rule. Variation in pain qualities is highest in CPSP and SCI CP. Attempts to
correlate various pain descriptors with some pathophysiology have failed. The most
common pain qualities appear to be burning, aching, lancinating, pricking, lacerating
and pressing (but also shooting, stabbing, squeezing, cramping, throbbing, tearing,
smarting, cutting, pulling, crushing, sore, splitting, icy feeling, stinging, like a tight
armor, sitting heavily on a ball, like a flash of lightning). Dysesthetic pain is
common in MS and incomplete SCI (including post-cordotomy), but, upon close
questioning, may turn out to consist of a number of specific pain qualities. A burning
quality is not a hallmark of CP, and in the landmark Danish paper (Andersen et al.
1995), lacerating was the commonest descriptor of pain. The more introspective
point out that their symptoms bear no relation to anything they have experienced in
the past. Whereas the majority have pain that can be described, several have no pain
at all, but an unpleasant and difficult-to-describe sensation that drastically reduces
their quality of life; moreover, there may be no sharp transition from non-painful to
painful dysesthesias. Some patients complain of pruritus, singly or in combination of

47

48

Central Pain Syndrome

TABLE 2.8. Pain quality

Muscle spindle pain

Burning

Cold

A cramp or contraction, with burning. There are sometimes areas of constant

cramping sensation, usually in a single muscle belly, as well as diffuse burning
when the muscle takes on a load. Weight-bearing while sleeping or resting on a
surface also causes great soreness, so that pts feels like they have been
sleeping on rocks. (This, plus the burning dysesthesia from touching bedclothes,
can make sleeping a torment.) Pts may describe muscle spindle pain as
drawing or pulling or crushing
A chemical, not a purely physical burn
Terms used:
mentholated burning
like the skin of my legs has been destroyed and the charred flesh turned up at
the edges
like in a dry lake bed
a sick burn, like that inflicted by a toxic chemical
a scalding, scathing torment, like in hell
like touching dry ice so that it burns
my hand tells me the skin of my legs is cold but it feels like burning
like I am touching an incredibly cold pipe in a freezing night, so that it drains the
flesh and burns me
like a dentist is touching the nerve in my tooth, only very cold

Metallic

like tinfoil under my skin

creepy, like chewing tinfoil

Wetness

When I am wet and sweaty, my skin is really sensitized and the burning lights up
and I feel wet and uncomfortable underneath the burning
I feel like I am being put on ice and then put into a fire with a million ice picks
plunged into my body (Bette Hamilton, one of Dr. Kevorkians clients): this
includes the burn, the cold, the metallic, and adds the lancinating component of
CP
often intolerable . . . crushed feeling, scalding sensation, as if boiling water
was being poured down the arm, cramping, aching, soreness, as if the leg was
bursting, something crawling under the skin, pain pumping up and down the
side, as if the painful region was covered with ulcers, as if pulling a dressing
from a wound, as if a log of wood was hanging down from the shoulder, as if
little pins were sticking into the fingers, like a wheel running over the arm, cold
stinging feeling (Head and Holmes 1911)
boiling hot, deep as though in the bones, showers of pain like electric shocks
or red-hot needles evoked by touch, as though the arm and leg were being
twisted, continuous sensation of pins and needles, a strange sensation of the
limbs being abnormally full (Loh et al. 1981)
as if knives heated in Hells hottest corner were tearing me to pieces (Holmes
1919)
Pins-and-needles

Dysesthesia in the aggregate

Circulatory
Visceral (peristaltic)
Pruritus

Burning in the bladder, fullness or nausea in the gut (like my bowels will
explode), heightened sense of distension and urgency with flatus or stool
This may occur singly or combined with other qualities

Central Pain of Brain Origin: Epidemiology and Clinical Features

other above-cited qualities (Table 2.9; Canavero et al. 1997). Paresthesias can also be
the main complaint. Numbness is experienced by many; it can occur with both total
loss of tactile sensibility, but also normal thresholds to touch, and sometimes
describes patients paresthesias or dysesthesias.
According to Dr. McHenry (www.painonline.org), himself a CP patient, patients
when asked to describe their pain quality sound like pain imbeciles and will only tell
of the components if they listen very carefully, and then only with cues from the
examiner. The result is that clinicians receive the false impression that CP is singular
when it is plural, especially in symptoms other than dysesthetic burning. The patient
of necessity borrows verbal descriptors from nociceptive pain, but these may mislead
the examiner, leading to conflicts that the patient cannot explain and decreased
credibility. Burning dysesthesia is an amalgam of pain sensations, but most closely
corresponds to the second pain that follows, for example, touching a hot stove. There
is nearly always a cold component, and frequently there is a metallic quality, as well as
a sensation of wetness.
10. Intensity of pain
Intensity varies widely between individuals, and severe pain is commoner among
paretics rather than plegics; the suicidal people are usually paretic. After
lenticulocapsular stroke, intensity tends to be maximal in the leg rather than in
the arm or face (Kim 2003). Generally speaking, CP tends to be worst in areas of
most severe initial sensory loss, while its evoked components are usually worst in
areas of retained or only mildly impaired sensibility. Globally, most patients consider
the pain to be severe or even excruciating, although some of them rate the pain
intensity rather low on rating scales. However, even when low or moderate, CP can
be assessed as severe because it causes much suffering and burden due to its irritating
character and constant presence. Pain can be assessed as a worse handicap than, for
example, severe motor impairment. It is difficult to say if intensity of pain is worse
with lesions at some levels rather than others, since published studies lack adequate
power. In our experience, there appears to be no meaningful difference among
suprathalamic, thalamic, brainstem or cord lesions. Intensity can be constant or more
often may fluctuate spontaneously, even paroxysmally, or following aggravating or
mitigating stimuli. Interestingly, variation in intensity may differ between pain
qualities in the same patient. In its more extreme, intractable form, the patient is
motivated to commit suicide. For most patients, the intensity of CP is sufficient to
interfere with daily activities and is a potential or active factor in the development of
depression, along with neurologic disabilities, themselves a risk factor; depression
may, in turn, increase the perceived intensity and affective quality of the pain.
11. Components
Patients with CP demonstrate three types of pain: (1) a constant spontaneous
component (almost all); (2) an intermittent (every day, with pain-free intervals
lasting a few hours at most), brief (seconds to minutes), intense, spontaneous
component (about 15%), generally shooting, shock-like or lancinating and with a
similar distribution to that of steady pain; when present, it can be the major
complaint; and (3) evoked pain (65%), that is, hyperesthesia, hyperpathia,

49

50

Hemorrhage Not
specified

Infarct

M 54

F64

5 of these pts
treated with
carbamazepine
or amitriptyline.
However, all 9 relieved
at 3 mo follow-up
(in F 67, amitriptyline
50 mg benadryl)
Hypodensity
in middle
cerebral
artery
territory
hypodensity
in internal
capsule
Hypodensity
in middle
cerebral

Not
specified

Not
specified

Infarct

M 36

Lessen
paroxysms
but pruritus
more
persistent
improvement

Phenytoin
cyproheptadine.
Carbamazepine

No lesion on
sensory
axis

Left limbs,
left trunk
(both
pruritus
and
hypesthesia)

Left
(hemisoma
?)
Right
(hemisoma)

Hemianesthesia

Left
(hemisoma)

Hemianesthesia

Pin
sensibility
decreased
Occasional
touch
allodynia
(perceived
as pruritus)

Not
specified

Massey
(1984)

Nocardia
abscesses
(2) F
(right) no
mass
effect

M 43

Reduction
in intensity,
frequency
and duration

Carbamazepine

Hypodensity
of posterior
limb internal
capsule/
lateral
aspect of
frontal and
temporal
lobes (right)

Effect

Drugs

CT/MR

left limbs,
left trunk
(both
pruritus
and
hypesthesia)

Sensory findings Site

Pain
Postoperative, temperature
hypesthesia
over
several
weeks

Sullivan and
Drake (1984)

SAH
(basilar
tip
aneurysm)
wrapping.
Also:
stenosis
90% right
ICA

Onset

F 58

Sex/age Trigger

King et al.
(1982)

Authors

TABLE 2.9. Central neurogenic pruritus due to brain lesions (excluding MS)

EEG: no focus
associated
to pruritus

EEG: mild diffuse

slowing irregular
polymorphic right
frontotemporal delta
rhythm. Itching
uncorrelated to EEG
changes. Further
complaints: painful fits

Also occasional
paroxysmal sensations
of warmth in the same
distribution of pruritus. EEG:
focus of intermittent slow
activity sharp
waves spikes in
frontotemporal region.
Episodes of pruritus
uncorrelated with
slow sharp activity

Notes

51

Shapiro
and
Braun
(1987)

Infarct

Hemorrhage Not
specified

Infarct

Infarct

Infarct

M 61

F 62

M 76

F 67

F 74

Days

Postoperative,
over
c.1 mo
(carotid
surgery)

Not
specified

Not
specified

Not
specified

Infarct

M 68

NS

Infarct

M 72

Normal,
except
poor 2-point
discrimination
(on left)

Hemianesthesia
(pruritus
bilateral
worse
on left)

Pinprick
sensibility
decreased
(left
hemisoma)

Hemianesthesia

Hemianesthesia

left ear,
cheek,
ala nasi,
upper lip,
neck, upper
back, knee

Left
hemisoma

Right
(hemisoma
?)
Left
forearm
and leg
(pruritus)

Left
(hemisoma
?)
Right
(hemisoma
?)
Left
(hemisoma
?)

Hypodensity
(superficial)
in parietal
lobe

artery territory
(parietal)
Hypodensity
in internal
capsule
Hypodensity
in internal
capsule
Hypodensity
in middle
cerebral artery
territory
Hypodensity
in internal
capsule
Hypodensity
in internal
capsule
middle
cerebral
artery territory

focal
Hypodensity
in middle
cerebral
artery
territory
internal
capsule
Amitriptyline
(20 mg/day)

Significant but
incomplete

spontaneous
disappearance

(continued)

52

Right
thalamic
stroke

Right
MCA
stroke

F 74

M 69

KimyaiAsadi
et al.
(1999)

Left-sided
hemiplegia

Normal (?)

Several
weeks

Several
days

Not
available

Hyperpathia

Vuadens
et al.
(1994)

F 69

Cavernoma
at T1 (MRI)

Not
specified

Dysesthetic
area inner
aspect
right arm

6 yr long
pruritus
also aching
pain
Itch appeared
late and
preceded CP
by at least
4 yrs

Various
localized
areas of
the left
trunk and
extremities
Left thigh

Left nose
and
throat

Bilateral
(starting
on the left)

Sensory findings Site

2 weeks

Not
specified

Onset

Central neurogenic pruritus due to spinal cord (excluding MS) lesions

SAH

M 37

Canavero
et al.
(1997)

F 82

Sex/age Trigger

Procacci and
Maresca
(1991)

Authors

TABLE 2.9 (continued)

54 drugs incl.
amitriptyline
at full dosage.
Propofol test.
IT baclofen
Topical therapies
(moisturizers,
emollients)

Not available

Not specified

Amitriptyline
Infarction of
(50 mg/day)
the territory
of the middle
cerebral artery

Right
thalamic
stroke

Ineffective

Antihistaminics
psychotropics

Negative
(also at MRI)

Episodic pruritus.
Right side spared.
Oral medications
refused

Intense pruritus, 2 yr
long, worse in the
morning

Notes

Not specified

Localized, unremitting
Effective
(or spontaneous pruritus, interfering
with sleep. Pruritus
resolution?)
resolved in a week

Ineffective.
Diazepam
10
25 mg
transitorily
effective
Alleviation of
each episode
of pruritus

Effect

Drugs

CT/MR

53

M 54

Dey et al.
(2005)

Cavernoma
at C3
4

Cavernoma
at T9
10

Gradual

Sudden

Unilateral,
focal itch
(after pain)

Pain plus
intense
itch; then
pain abated,
and itch spread
Itch appeared
on the 9th year
of symptoms
(pain)
Moderately
relieved itch

Improved
pain but
not itch

Itch
unrelieved

Worsened
the itch

Lidocaine
patch 5%,
EMLA cream
and gabapentin
Opioids

TCAs, SSRIs,
AEDs (OXCBZ,
CBZ, zonisamide,
tiagabine,
levetiracetam),
IV lidocaine,
stellate ganglion
block with lidocaine
TENS and
acupuncture

Excision at first Normal

completely
postop
relieved both
MRI
CP and itch.
3 mos postop,
neck and
shoulder pain
recurred,
radiating down
left arm to base
of left thumb,
spreading over
2 yrs to whole
hand. Pain
changed from
intermittent and
stabbing to
constant and
burning. Itch
recurred 2 yrs
postop

Marked relief
No response
to H1 blockers
and steroids
Topiramate
ineffective on
both itch and CP
Other AD/AED
ineffective

Lidocaine
patch 5%

Mid-back (itch)
Groin (pain);
spread to whole
lower abdomen
below T9

Both itch and pain

improved by
distraction. Scratching
temporarily relieved
itch although
worsening the pain.
Some itch was felt
deep within the biceps
area of upper arm;
temporary relief
without pain
exacerbation by
squeezing biceps

Previous episodes of
typical CP in affected
hypesthesic areas,
each spontaneously
regressed

MS: Multiple sclerosis

The undetailed report by Andreev and Petkov describes pruritus of the nostrils (6 pts) as almost pathognomonic of a brain tumor infiltrating the base of the fourth ventricle (see in Canavero et al. 1997)
Another report of localized pruritus appears in Johnson et al. (2000) (reference not available).

In all pts other causes of pruritus were excluded by thorough investigation. The only dermatological findings were due to scratching. Pruritus in all came in paroxysms or bouts, intermittent or continuous

F 55

Sandroni
(2002)

54

Central Pain Syndrome

hyperalgesia and/or allodynia (Tasker 2001). Any single patient may, however,
complain of only one of these three components. Only a minority of CPSP patients
has their spontaneous CP absent for up to a few hours each day.
Shooting (lancinating) pain is the most distinct, most severe and most startling,
but it does not cause the most suffering, because the pain is limited to the surface
area affected and can often be eliminated by shifting position or rubbing the area;
this pain shoots from distal to proximal sites. The phenomenon is most dramatic
early in the disease and tends to diminish with time, leading to false notions of drug
benefit. It is indistinguishable clinically from the lightning pains of tabes dorsalis.
Lancinating pain is said to originate where mini-fasciculations occur (Dr McHenry,
painonline.org).
Paretics display the greatest number of CP components, unlike plegics and MS
patients (although the ones they have can be severe). Gradients can be observed,
namely spontaneous pain tends to be distal (i.e., where sensory loss becomes greatest)
and evoked pains proximal (i.e., where sensory loss is present but least marked).
12. Evoked pains
The spontaneous discomfort of CP is often (roughly 60%) accompanied by certain
unpleasant effects induced by somatosensory stimuli, and which, by definition,
cannot occur in an area of complete somatosensory interruption; it is unusual in the
complete absence of clinically detectable sensory loss (about 7% in Tasker et al.
1991). Infrequently, these can be the only symptoms, i.e., in the absence of constant
pain (3/27 in the series of Shieff (1991) and 7% in the series of Tasker [2001]), and
may first be noticed after several years with the disease. These abnormally unpleasant
sensations (pain, dysesthesias, paresthesias) are usually unbearable and evoke violent
emotional and defensive reactions, generally being referred to as the worst
component of CP; often poorly localized, they may be elicited either by normally
non-painful stimuli, namely touch (including caresses and others; see below)
but
not, at least initially, deep pressure
vibration, moderate cold and heat (allodynia)
or by mildly to moderately painful stimuli, particularly sharp objects plus noxious
cold and heat (hyperesthesia: hyperalgesia and hyperpathia) delivered to an area of
nearly (but not) always elevated threshold to stimuli of one or more somatosensory
modalities (thermal, mechanical static and dynamic). Hyperalgesia may be less
frequent in brainstem CP. These evoked pains are elicited most prominently by
a single sensory modality, a little more often than by several (Tasker 2001). Riddoch
(1938) and others noted how pain can be evoked by simple pressure in areas of
analgesia to pinprick. Also, even in the presence of nearly abolished pinprick
sensibility, firm pinching or repeated pinpricks may be felt as painful. Head and
Holmes (1911) also noted how pressure (deep tissue pain) with an algometer could
evoke discomfort in cases with complete analgesia to pinprick (as reemphasized by
Mailis and Bennett 2002). In patients with complete thermanesthesia, extremes of
heat and cold may evoke disagreeable nonthermal sensations (Riddoch 1938).
Allachesthesia is allodynic pain in an area other than that stimulated. According to
the IASP nomenclature (1994), hyperpathia (a term first introduced by Foerster) is
a painful syndrome, characterized by an abnormally painful reaction to a stimulus,
especially a repetitive stimulus, as well as an increased threshold. Riddoch (1938)

Central Pain of Brain Origin: Epidemiology and Clinical Features

(wrongly), but also Head and Holmes (1911), considered these to be the cardinal
feature of CP: The sensation evoked is abnormal. The painful sensation develops
explosively. There is usually little relation between the strength of the stimulus and
the amount of sensation excited: it is nearly all or nothing. Moreover, there is no
refractory period for hyperpathic responses. The effective stimulus may include all
somatosensory stimuli or only a specific type of input (such as cold or draft, the light
touch of clothing or pinprick, even smoke). These grossly unpleasant sensations may
demonstrate temporal or spatial spread.
Simple neurologic sensory tests characterize radiation of pain or dysesthesia (to
body areas not directly in contact with the pain-evoking stimulus: in a hot room . . .
if one rubs the whiskers of the face with the palm of the hand, burning is felt in
the ulnar forearm. Sitting on a chair until the burning is prominent on points of
contact, burning is also felt in the lateral thigh which is not in contact with the
fabric of the chair), present in half the cases, after-sensations (the persistence of
pain long after the stimulus and the arrival of primary afferent impulses that evoke
pain), seen in about 40% of cases, and prolonged temporal summation (the gradual
build-up of pain with repeated stimulation) (Garcin 1937; Riddoch 1938). Radiation
of sensations from the stimulus site and spatial and temporal summation appear
to be more common in CP than in PNP. Although response latencies can be normal,
anomalous summations may be seen: slow temporal (pain or dysesthesias start
after a delay which, during the daytime, the patient can anticipate and avoid: if
occlusive touch is applied to the skin, within minutes, evocation of the spontaneous
dysesthetic burning occurs. The stimulus may be roughness, but the patient perceives
it as heat. The search for ever cooler shoes may be launched when what is needed
is smooth leather, not the sueded tongue which is common), very slow temporal
(starting after hours: as to confinement or weight-bearing it renders a nights
recumbency as feeling like the bed was hard as rocks. As to exercise, it means
the muscle soreness the day after exertion is overwhelming), delayed with overshoot
(this is not a temporal delay: rather it is a heightened threshold for pain, which,
when reached, overshoots wildly, most easily seen in the response to sharp objects.
A normal will note graded sharpness as painful before a CP patient will, but because
the pin in pinprick testing is so sharp, this delay is often missed at examination),
spatial (an unexpected increase in pain as the area of stimulus is increased: it appears
never to have been tested in CP). Wind-up pain (increasing pain with increasing
numbers of pinpricks, i.e., temporal summation) has been reported in CP (see also
Bowsher 2005).
In other words, provoked pain is characterized by late onset and poor localization,
generally radiates from the stimulated point to the entire half of the body (one third
of cases) or lesser body areas and persists for an unusually long time after stimulation
has ceased. Evoked pains have a distribution which is less widespread than that of
steady or intermittent pain. As a rule, somatic stimuli can cause or aggravate pain
only when applied to the affected side, but sometimes even the stimulation of the
normal side gives rise to exacerbation of pain (synesthesalgia).
Patients may wear as little clothing as possible over affected areas and seek a
narrow window of room temperature, or alternatively wear gloves to avoid contact
with the painful hand.

55

56

Central Pain Syndrome

Paradoxical burning on cold stimulation is reported by some patients with CP

(e.g., Berglund et al. 2001; Bowsher 2005). According to the review of Greenspan and
colleagues (2004, and references therein), all studies report a large proportion of CP
patients with cool (about 50%) or warm hypesthesia, with no more than 23% (but
50% in the series of Attal et al. 2000 and 56% in that of Andersen et al. 1995) showing
cold allodynia (see also Morin et al. 2002) and very few or no cases of heat allodynia
(e.g., 3/16 in Attal et al. 2000). In their personal series, two patients with bilateral
warm hypesthesia also had bilateral cold hypesthesia, with same-side prevalence; cold
allodynia occurred uncommonly among patients with cold hypesthesia (2/11), both
unilateral CPSP, who also had bilateral cold hypesthesia. Interestingly, the patient
with normal cold detection threshold had the most extreme cold allodynia (in this
case, cold allodynia was evoked at temperatures cold enough to activate receptors in
the cool pathway, but not those of the supposed heat-pinch-cold pathway). Tactile
allodynia is reported by about 40% of patients. Hair sensation is usually unaffected
and has never been reported to cause burning.
A review of all published cases and case series of CP over a century shows CP
exacerbation by environmental changes (wind, weather changes, low atmospheric
pressure, altitude, cold or warm temperatures), emotional stress (sudden fear, joy,
anxiety, depression, others), tiredness, smell, loud noises, sad or distasteful music,
(sudden) bright light, movements (including vibrations and changing position) and
physical activity (e.g., walking, non-strenuous activity, isotonic
isometric muscle
contraction of one or more muscles together, with ensuing activation of muscle
stretch receptor afferents: this so-called movement/kinesthetic/proprioceptive
allodynia, seen in about 10
20% of patients, can hinder rehabilitation and virtually
paralyze some patients), visceral stimuli (e.g., a full urinary bladder or rectum,
drinking cold and warm water, passing urine, cough, Valsalva maneuver), the
thermal grill, smoking (and even the curling of cigar smoke along the fingers),
intellectual concentration, inactivity (such as attempts to sleep), merely blowing on
the skin and combing the hair. Less commonly, similar stimuli may reduce the pain.
Dyskinesias and other anomalous motor reactions can also worsen CP. Rarely, an
overresponse to pleasant stimuli or relief by pleasant stimuli (e.g., warmth or
orgasm) may also be found (Riddoch 1938); for instance, Biemond (1956) described
a patient who drew a passing sensation of pleasure with cold drinks and ice creams.
Table 2.10 gives a summary of studies focusing on all discussed clinical features.
13. Somatosensory findings (Tables 2.10
2.12)
Dejerine and Roussy (1906) concluded that persistent loss of superficial sensation
(hemianesthesie superficielle persistante) to touch, pain and temperature, associated with a more pronounced and persistent loss of deep sensibility, was typical
of thalamic pain. Sometimes, superficial hemianesthesia was replaced by hyperesthesia. They also noted a predominantly distal hemianesthesia (or hypesthesia),
less pronounced proximally on the limbs, slightly exceeding the midline. Pain and
thermal sensibility were reduced, but not totally abolished. Patients could not
recognize the nature of the stimulus and the site of the stimulation and complained
of dysesthesia, topoanesthesia and topoanalgesia, with a delayed perception of the
stimuli. In many cases these troubles of superficial sensibility were subtle,

Central Pain of Brain Origin: Epidemiology and Clinical Features

57

58

Central Pain Syndrome

Central Pain of Brain Origin: Epidemiology and Clinical Features

59

60

Central Pain Syndrome

Central Pain of Brain Origin: Epidemiology and Clinical Features

61

62

Central Pain Syndrome

Central Pain of Brain Origin: Epidemiology and Clinical Features

63

64
R
0
nt
nt

0
0

nr
nr

51
L


0

nr

nr
nr

Age

Side of pain

Tactile sensibility (Von Frey)

2-point discrimination

Localization of stimuli

Threshold for prick

Unpleasant response to prick

Threshold for painful pressure

Unpleasant response to pressure

Sensibility to heat

Sensibility to cold

Unpleasant response to extreme heat

Unpleasant response to extreme cold

Pleasant response to mild warmth

Unpleasant response to visceral stimulation

Unpleasant response to tickling/scraping

Appreciation of vibration

nr

(shifting)

(shifting)

59

nr

nr

nr

(sole)

64

60

/0

nr

(shin)

(hand)

(palm, sole)

nt

nt

65

10

nr

nr

/

52

11

nr

nr

(sole)

nt

(head)

65

12

/0

nr

(sole)

nt

nt

43

13

Head and Holmes objectively analyzed sensory loss and dissociation of sensibility in pts with lesions of the CNS at spinal, mid-brain, thalamic and cortical level by means of an instrumentation that in
some cases was designed expressly for this purpose. Results on the affected part were always compared with results obtained in the unaffected similar part of the body. Data were recorded as accurately
and objectively as possible. Light touch was examined first by applying a wisp of fine cotton wool avoiding any deformation of structure. For determining the threshold for light touch the authors employed

0, lost; , unchanged (no difference between affected and unaffected side);
, diminished;

, strongly diminished; , increased; , strongly increased; nt, not tested; nr, not reported.

nr

nr

49

Sex

Case no.

TABLE 2.10a. Somatosensory troubles in Head and Holmes cases of thalamic pain

65

(continued)

As of loss of superficial and deep sensibility, in some pts with thalamic syndrome this loss is so insignificant that it can be discovered by measurement only, so we can imagine the existence of the
over-reaction without it. Even if all pts with over-reaction had a more or less recognizable sensory loss, the excessive response bears no relation to the extent of the accompanying loss of sensation.
They noted that the appreciation of posture and recognition of passive movements is impaired more frequently than any other sensory quality. The amount of this loss varies from a scarcely measurable
defect to a complete loss of these sensibilities.
Tactile sensibility was frequently diminished and in some cases totally lost, but generally a threshold could be obtained, especially increasing the strength of the stimulus. Tactile threshold, measured with
Von Freys hairs, was unchanged between the two halves of the body in 5 cases, but in the majority of the cases it was raised on the affected side. In few cases only the affected parts were totally
insensitive to the tactile hairs and also to pressure-aesthesiometer. In some pts, the consecutive contacts (especially with increasing strength) caused widespread tingling that made conclusive
demonstration of the threshold impossible. Determination of tactile threshold was also prevented by the occurrence of involuntary (induced) movements, with accessory sensations misinterpreted as
stimulation.

Von Freys graduated hairs ranging from 8 to 110 g/mm2. They always performed 16 contacts in 1 minute, avoiding rhythmicity. The series of tests were performed without word exchange; hallucinatory
responses were also recorded. Pressure-touch was tested by contact with the observers finger provided that its surface temperature was similar to that of the part to be examined. The threshold for
pressure-touch was determined by a pressure-aesthesiometer. Specific and as accurate as possible methods were used to test the faculty of localization, the threshold for the appreciation of roughness,
the ability to discriminate two simultaneous contacts, the power of recognizing the posture of any part of the body, the power of appreciating passive movements and the power of appreciating weight,
size, bi-dimensional shape, three-dimensional form, texture and consistence of objects. The power of recognizing vibration was tested by means of a tuning-fork, beating 128 Hz, also noting the duration
of the sensation. Tickling and scraping were employed to evaluate the affective component of sensation.
Superficial sensibility to pain was tested first by pricking with a sharp steel pin or needle and a comparison between normal and affected parts was always performed. Being well aware that this test was
subject to a source of error due to the reduction of the power of recognizing the size (sharpness) of the stimulating object, in cases with slight disturbances of pain sensation, they determined the
threshold for pain by means of an aesthesiometer (algesimeter). They also noted that if a pain-spot was not directly stimulated, the same pressure was reported as touch. Finally, pressure-pain was tested
by means of a Cattell algometer, measuring the amount of pressure (kg) on a standard area necessary to evoke pain. Results of the test on the affected part were always compared with results in the
similar unaffected part of the body.
The thermal sensibility was examined by means of silver tubes filled with hot or cold water. The temperature of the water at the moment of testing was read on a thermometer. The authors determined the
threshold for heat and cold on similar parts of the two halves of the body, as well as the power to distinguish the relative warmth or coldness of two tubes. Moreover, the sensation evoked by neutral
temperature was compared with that of a distinctly cold or warm tube. They also observed the effect of extreme heat (50C or more) and cold (15C or less) and compared the sensation evoked on
normal and abnormal parts of the body. To study the affective component of thermal stimuli, they employed large glass tubes (4 cm in diameter) filled with water at various temperature. They also noted
that the temperature tests were liable to lead to erroneous conclusions due to the tendency to call all sensations evoked during the testing either hot or cold. Pts with thalamic lesions and capable of no
thermal appreciation were more liable to call hot every thermal stimulus and even repeated pricking. This confusion was more likely to occur in pts with over-response to affective stimuli. In many pts it
was also difficult to determine the extent of the neutral zone between heat and cold threshold, as pts possessed no word which expressed this neutral sensation (nothing but a touch).
They reported data on one pt with SCI (Brown
Sequard paralysis) without CP, 3 cases of brainstem lesion (1 of them with CP following a Wallenbergs syndrome), 9 cases of thalamic lesion (thalamic
syndrome) and 5 pts with cortical lesions (1 of them reporting pain during sensory epileptic attacks). Their conclusions on neurological features in thalamic syndrome were however based on data on 24
pts. In their opinion, the essential feature of thalamic syndrome is the tendency to react excessively to unpleasant stimuli (over-reaction).
In the pt with CP following Wallenbergs syndrome the sensibility to light touch (cotton-wool, Von Freys hairs), the appreciation of roughness (Graham
Brown aesthesiometer) or of two simultaneous
contacts and the power of recognizing vibration were not different between the two sides of the face, even if the pt complained that all forms of touch were less vivid over the affected (right) side. The
affected side of the face was insensitive to superficial pain (prick), but pressure-pain was not lost (the Cattell algometer gave approximately equal readings on the two sides). Both heat and cold were
appreciated on the two halves of the face and the thresholds were the same, but heat seemed hotter over the affected side while cold seemed less cold. On the body there was no difference in
appreciation of touch, roughness and vibration but sensations were more vivid on the normal (right) half of the body. Heat and cold could be appreciated, but heat seemed hotter on the affected (left) half
of the body and cold seemed less cold. The left half of the body, except an area in the left perineum, penis and scrotum, was insensitive to prick. The pressure of the algometer necessary to evoke pain
was considerably higher on the affected hemibody than in the normal half. The left testicle was insensitive to the pressure.
Somatosensory troubles in their pts suffering from thalamic syndrome are summarized in the table above.

66

The loss of thermal sensibility generally affected intermediate temperatures, yielding a sensation of pleasant warmth. However in several pts able to appreciate mild heat (34C), the application of water
at 38C on the affected part evoked a higher degree of pleasure than the same application over the unaffected part. In one case, excessive pleasure could be converted into excessive discomfort as soon
as water temperature exceeded 46C. In a few pts when thermal sensibility was abolished, warmth applied over a sufficiently large surface evoked a feeling of pleasure, even if the pt did not recognize it
was warm and extreme hot or cold evoked great discomfort.
Head and Holmes analyzed the effects of visceral stimulation in pts suffering from thalamic syndrome by comparing the effect elicited by squeezing testicles (without pinching the scrotum). They noted
that in many pts the discomfort was more intense and the cremasteric movements were more brisk after squeezing the testicle of the affected side. They also noted that even when pinprick pain threshold
on the glans penis were the same on both sides, the discomfort described by the pts was greater after pricking of the affected half.

Many pts (50% of cases) could not recognize the position of a stimulated spot. In many cases where tactile sensibility was diminished, the inability was maintained even with pricks or painful pressure, to
which the pt was sensitive. Pts could be at a loss where they were touched, or could refer touch to wrong areas. When the posture was not recognized and the power of localization was lost, pts recognized
the stimulus as a change within the part of themselves and did not refer the discomfort to the action of an external agent. Moreover, when localization was affected, unpleasant sensations could spread
widely over the affected part: for instance, they noted that a prick on the hand could cause a painful sensation in the cheek or side.
In no instance among 22 pts the threshold for pinprick pain was lower on the affected body side; it was identical on both sides in 13 cases and raised in 9 cases, in whom a stronger stimulus was needed
to produce a sensation of prick. Yet most pts (20/22) showed an over-response to prick.
They also attempted to measure the amount of pressure evoking pain, comparing the 2 sides of the body. They noted that the same pressure produced more disagreeable discomfort and increased
reaction on the affected side in every one of 24 pts tested. Moreover the pain developed explosively, as the pressure increased over a certain point. They noted that the threshold for pressure pain was
frequently lower on the affected side of the body (15 cases), but it was higher in 3 cases and unchanged in 6 cases. No pts showing a lowered threshold for painful pressure did show a lower threshold for
pinprick pain. Yet, the response on the affected half of the body was excessive in all 24 pts. They also stated that excessive pressure (especially on a bone) normally caused discomfort rather than pain,
and that distressing sensation differs profoundly from the pain produced by a prick, even if both stimuli were perceived as painful. They concluded that pressure pain contain some sensory factors to
which the affected half of the body is peculiarly susceptible and the over-reaction was due to this increased susceptibility, rather than increased sensibility to pain (as demonstrated by the fact that
threshold to pinprick may be raised in pts with lowered threshold to pressure). A reduced sensibility to pain delays the appearance of the over-reaction, but, as the stimulus is strong enough to cause
pain, the discomfort greatly exceeds that produced over the unaffected part.
Sensibility to heat and cold could show all degrees of change from total loss to a slight increase of the neutral zone. Thermal appreciation could be unaltered, even if, in the majority of cases, it was
diminished or lost and threshold for the appreciation of heat and cold were never lowered and could be the same on both sides of the body or be raised. The threshold for thermal stimuli and the range of
discrimination was normal and was the same on the 2 sides of the body in 37.5% of cases (9/24 pts). In these pts could appear an over-response to pleasurable heat. However, in many cases, the
appreciation of heat and cold was abolished and ice and water over 50C evoked only discomfort on the affected side. The threshold for heat-induced over-reaction was about 40
45C in most pts, but
in some cases temperatures of 55
60C were needed. The threshold for cold-induced over-reaction was generally below 15C. The evoked sensation was the same whichever of the two extremes was
used and the pt cannot recognize the cause of the unpleasant sensation. This over-reaction could occur both in pts in whom the threshold for the appreciation of heat and cold was identical on the two
halves of the body and in cases where the sensibility to heat and cold was completely lost: 22 out of 24 pts with thalamic syndrome showed an excessive response to extreme heat and cold. Even though,
in many cases, only heat above 50C and cold below 15C (or melting ice) evoked the over-reaction, in some cases with lesser thermal serangement, temperatures below 26C and above 40C evoked
this indiscriminate response from the affected half of the body.
Some pts, with less severe impairment of thermal sensibility, could recognize temperature above 38
40C as warm and those below 26
28C as cold. However, any temperature that could be
appreciated was hotter or colder on the affected side than on the unaffected one, perhaps due to the increased affective reaction.
They also stated that it was unlikely that pts had an actual increase in sensibility to temperature, but they simply translated the increased discomfort into terms of greater cold or heat; moreover, in pts
suffering from thalamic lesions, the power of appreciating either heat or cold cannot be lost singly.
In other words, heat and cold are not dissociated; if one form of sensation is lost, the other will be gravely disturbed.

TABLE 2.10a (continued)

67

A highly educated patient confessed that he had become more amorous since the attack, which had rendered the right half of his body more responsive to pleasant and unpleasant stimuli. I crave to
place my right hand on the soft skin of a woman. Its my right hand that wants the consolation. I seem to crave for sympathy on my right side. Finally he added, My right hand seems to be more artistic.

We were anxious to discover if sensations, normally accompanied by a pleasurable feeling-tone, also produced a similar over-reaction. Unfortunately, the greater number of methods . . . either produce
discomfort or . . . an entirely indifferent sensation. But in the milder degrees of heat we possess a measurable stimulus (!!) endowed with a pleasant feeling-tone . . . In a few cases when thermal
sensibility was abolished, warmth applied over a sufficient large surface evoked a feeling of pleasure . . . One of our patients found a hot-water bottle pleasant and soothing to the affected foot, but did
not recognize that it was warm until he touched it with some normal part . . . Many patients found the warm hand of the observer unusually pleasant on the abnormal side, although no such
manifestations of pleasure were produced when it was applied to the normal part of the body. In one case . . . the patient could not recognize any thermal stimulus as such, and yet over the affected half
of the chest . . . water at from 38C to 48C evoked intense pleasure. Temperature of 50C and above, or of 18C and below, caused great discomfort [three cases are described and several patients
referred to]. So far we have been unable to find any temperature which produces a sensation of pleasurable cold.
Behavior of the affected half of the body in states of emotion (p. 135):

Pts complaining of thalamic pain could complain of unpleasant sensations after scraping the palm or the sole of the foot, or moving a rough object over the skin or even rubbing the hairs over the affected
part of the body. Sometimes, these sensations were not painful, but very unpleasant and frequently they spread from the stimulated area to the entire limb or half of the body. Examination with a
Graham
Brown aesthesiometer (to estimate the appreciation of roughness) frequently induced this anomalous response. Nevertheless, the threshold of appreciation of roughness was never lowered. It
was always unchanged or increased, but in the large majority of the pts the aesthesiometer induced greater discomfort on the affected side. Occasionally even the vibration of the tuning fork was able to
give rise to similar spreading sensations. In pts characterized by an over-response to painful stimuli tickling was also unpleasant and induced greater reaction.
The vibrations of a tuning fork were generally appreciated on both halves of the body, but in almost every case for a shorter time on the affected side. In many cases, the pt complained that vibrations
were not so plain or tuning fork vibrated less rapidly on the affected side. Only in few cases (in whom most other sensations were gravely affected), the affected half of the body was insensitive to this
stimulus. They noted that a shortened appreciation of the vibration of a tuning fork is associated with the over-response to painful stimuli, independently of the unpleasant feeling-tone evoked by
vibration.
Response to pleasurable stimuli (p. 133):

68

Central Pain Syndrome

TABLE 2.10b. Review of somatosensory abnormalities in Head and Holmes published

cases, Head and Holmes (1911)
24 CP pts.
As far as the loss of superficial and deep sensibility is concerned, in some patients with thalamic syndrome this loss is
so insignificant that it can be discovered by measurement only, so we can imagine the existence of the over-reaction without
it. Even if all pts with over-reaction had a more or less recognizable sensory loss, the excessive response bears no relation
to the extent of the accompanying loss of sensation
Tactile threshold (von Freys hairs):
identical on both sides: 5/24 pts (20.8%)
raised or lost or undetermined on the affected side: 19/24 pts (79.2%) *
Tactile sensibility was frequently diminished and in some cases totally lost, but generally a threshold could be obtained,
especially increasing the strength of the stimulus. In few cases only the affected parts were totally insensitive to the tactile
hairs and also to pressure-aesthesiometer
Threshold for pinprick pain:
identical on both sides: 13/22 pts (59.1%)
raised on the affected side (a stronger stimulus was needed to produce a sensation of prick): 9/22 pts (40.9%)
lower on the affected side: 0/22 pts (0%)
Over-response to prick: 20/22 pts (90.9%)
Threshold for thermal stimuli and range of discrimination:
raised on the affected side: 15/24 pts (62.5%)
normal and identical on both sides: 9/24 pts (37.5%)
lower on the affected side: 0/24 pts (0%)
Sensibility to heat and cold could show all degrees of change from total loss to a slight increase of the neutral zone.
Thermal appreciation could be unaltered, even if, in the majority of cases, it was diminished or lost. The loss of thermal
sensibility generally affected intermediate temperatures.
In pts with normal threshold could appear an over-response to pleasurable heat. In pts with abolished appreciation of heat
and cold, ice and water over 50C evoked only discomfort on the affected side. In pts suffering from thalamic lesions, the
power of appreciating either heat or cold could not be lost singly. In other words, heat and cold are not dissociated; if one
form of sensation is lost, the other will be gravely disturbed.
Threshold for heat-induced over-reaction:
about 40
45C in most pts (55
60C in some cases)
Threshold for cold-induced over-reaction:
generally below 15C
The evoked sensation was the same whichever of the two extremes was used and the pt could not recognize the cause of
the unpleasant sensation
Threshold for pressure pain:
lower on the affected side: 15/24 pts (62.5%)
identical on both sides: 6/24 pts (25%)
raised on the affected side: 3/24 pts (12.5%)
Visceral stimulation (comparison of the effects elicited by squeezing testicles without pinching the scrotum):
In many pts the discomfort was more intense and the cremasteric movements were more brisk after squeezing the testicle of
the affected side. Even when pinprick pain threshold on the glans penis were the same on both sides, the discomfort
described by the pts was greater after pricking of the affected half
Vibrations of a tuning fork (128 Hz):
Generally appreciated on both halves of the body, but in almost every case for a shorter time on the affected side;
vibrations not so plain or tuning fork vibrating less rapidly on the affected side. Only in few cases (in whom most other
sensations were gravely affected) the affected half of the body was insensitive to this stimulus. A shortened appreciation of
the vibration of a tuning fork was associated with the over-response to painful stimuli, independently of the unpleasant
feeling-tone evoked by vibration
*

In some pts the consecutive contacts (especially with increasing strength) caused widespread tingling that made
conclusive demonstration of the threshold impossible. Determination of tactile threshold was also prevented by the
occurrence of involuntary (induced) movements, with accessory sensations misinterpreted as stimulation.

69

49

Age

Side of the cortical lesion

(hand, ulnar)

nr

nr

nr

nr

nr

Hyperesthesia (over-reaction)

Allodynia

Localization

Sense of position

Vibration

Site of the lesion (autopsy)

Thalamic lesion(s)

Cortical/subcortical
fronto-parietotemporo-occipital
N

0, lost; , unchanged (no difference between affected and unaffected side);
, diminished;

, strongly diminished; , increased; nt, not tested; nr, not reported.

nr

nr

Parieto-temporoinsular

nt

nr

nr

(hemibody)

nr

Fronto-parietotemporo-occipital
insula
N

nr

nr

(hemibody)

nr

nt

(foot)

(foot)

nr

Foix

(hemibody)

(hemibody)

(hemibody)

53

Davidson/Schick

nr

Y (hemibody)

0 (hemibody)

(hemibody)

(hemibody)

nr

Roussy/Foix

nr

nr

(cheiro-oral)

(cheiro-oral)

Pinprick sensibility

Thermal sensibility

(hand, ulnar)

(hand, radial)

nr

De Ajuriaguerra/Lehrmitte

Tactile sensibility

Somatosensory troubles contralateral to the lesion

Pts sex

Lehrmitte/Claude

TABLE 2.11. Somatosensory troubles in the review of De Ajuriaguerra (1937)

70

Central Pain Syndrome

Central Pain of Brain Origin: Epidemiology and Clinical Features

71

72

Central Pain Syndrome

Central Pain of Brain Origin: Epidemiology and Clinical Features

73

74

Central Pain Syndrome

Central Pain of Brain Origin: Epidemiology and Clinical Features

75

76

Central Pain Syndrome

Central Pain of Brain Origin: Epidemiology and Clinical Features

77

78

Central Pain Syndrome

Central Pain of Brain Origin: Epidemiology and Clinical Features

79

80

Central Pain Syndrome

Central Pain of Brain Origin: Epidemiology and Clinical Features

81

82

Central Pain Syndrome

Central Pain of Brain Origin: Epidemiology and Clinical Features

83

84

Central Pain Syndrome

Central Pain of Brain Origin: Epidemiology and Clinical Features

85

86

Central Pain Syndrome

Central Pain of Brain Origin: Epidemiology and Clinical Features

87

88

Central Pain Syndrome

Central Pain of Brain Origin: Epidemiology and Clinical Features

89

90

Central Pain Syndrome

Central Pain of Brain Origin: Epidemiology and Clinical Features

91

92

Central Pain Syndrome

Central Pain of Brain Origin: Epidemiology and Clinical Features

93

94

Central Pain Syndrome

Central Pain of Brain Origin: Epidemiology and Clinical Features

95

96

Central Pain Syndrome

Central Pain of Brain Origin: Epidemiology and Clinical Features

97

98

Central Pain Syndrome

necessitating an accurate neurological examination to detect them. Deep sensibility

(articular, muscular, tendineal and osseous) was more profoundly and persistently
affected. In many cases, patients could neither recognize vibration nor perceive
active and passive movements, muscle power and strength; there was also a loss of
joint position sense. Dejerine and Roussy noted in their patients the presence of
allodynia, hyperesthesia and hyperalgesia/hyperpathia (described as excessive
reaction to touch, cold or warm and pinprick). Head and Holmes (1911) later
published an unsurpassed quantitative clinical analysis of sensory abnormalities in
CP patients. Ever since, we know a wide spectrum of sensory abnormalities can be
found among patients with CP. They range from a slightly raised threshold for
one of the submodalities, to complete loss of all somatic sensibility in the painful
region, or a very painful hyperesthesia. In some patients the abnormalities are subtle,
but can often be detected by quantitative sensory tests (QST), as demonstrated by
Head and Holmes.
A survey of the literature shows that the common feature of more than 95% of
all CP patients is impaired temperature and pain (i.e., spinothalamic) sensibility

Central Pain of Brain Origin: Epidemiology and Clinical Features

at clinical or electrophysiological examination (Garcin 1937; Riddoch 1938;

Tasker 2001); appreciation of pinprick and temperature is nearly always impaired,
and there is almost always a raised threshold to innocuous thermal (both warm and
cold) detection and to a lesser extent also to painful heat and pain. Some patients
who have lost the ability to perceive cold and warm due to CNS lesions can
nonetheless distinguish warm or cool objects by the distinctly different feelings they
evoke (e.g., Kinnier Wilson 1927; Davison and Schick 1935). No unequivocal report
of CP arising from lesions restricted to the lemniscal pathways has been published,
and several patients (particularly in Wallenbergs syndrome) have normal thresholds
for touch, vibration and kinesthesis (in such cases, the posterior columns may
mediate evoked pains); instead, many cases of lesions restricted to the STT are
on record (cordotomy, anterior spinal artery syndrome, medullary stroke). CP is
independent of other neurological symptoms, including paresis, tremor, dystonia,
speech disturbances, hemianopsia; only somatosensory abnormalities are always
present, although these are far from uniform among patients (see also Gonzales
et al. 2001). Pain distribution is usually well correlated with sensory abnormalities.
The pain may also occur in patients with brain lesions who have recovered from
clinically detectable sensory loss and persist in time; in this case, a crude sensory
examination, weeks or months after the lesion, reveals no sensory deficit.
Nonetheless, a lesion affecting the STT system is a necessary but not sufficient
condition for the development of CP.
It is the experience of all groups doing research with CP that a few patients do
not display thermoalgesic abnormalities (e.g. Garcin [1937]) reviewed cases in which
there seemed to be only loss of epicritic sensibility and De Ajuriaguerra [1937]
observed three patients without thermoalgesic deficits), even at QST. Examples
include Boivie and Leijon (1991; 1 case), the series of Tasker et al. (1991; 2 cases,
although one showed abnormalities of the late components of somatosensory evoked
potentials), Shieff (1991; 4 cases), Gonzales and colleagues (1992; 1 patient), Bowsher
(1996; 1 case
see other examples in Chapter 3, Section 4). Tasker and colleagues
(1991) described a patient with pain associated with cord lesions that caused
a preferential loss of touch, position and vibration with almost complete sparing
of STT. Regev and colleagues (1983) reported a patient with a pontine hemorrhage
producing no detectable somatosensory deficit, who developed transient spontaneous pain and allodynia to touch, pinprick and temperature. Sandyk (1985)
reported a CP patient with a cortical parietal hematoma whose only somatosensory
finding was allodynia induced by thermal (warm/cold) stimuli. However, in all
those cases where no sensory loss was seen in the first place, imaging techniques suggest a central lesion appropriately located to damage the somatosensory
system.
14. Sympathetic and other signs and symptoms
Signs of abnormal sympathetic nervous system activity within the region of disability (i.e., focal distribution) may sometimes be present: cooler and vasoconstricted skin in the painful area, edema, hypo/hyperhydrosis (rare), altered
skin texture and color (mottled skin or livedo) (Garcin 1937; Riddoch 1938).

99

100

Central Pain Syndrome

However, these signs are equally present in non-CP patients with CNS injury;
decreased movement alone can cause autonomic changes. A cerebral lesion can
cause trophic disturbances in contralateral limbs (Arseni and Boetz 1971),
particularly the shoulder hand syndrome, even paroxysmally (Montgomery and
King 1962). A common source of pain after stroke is nociceptive pain localized
to the shoulder and resulting from paresis and changed muscular tone/posture
and sensory loss. One fourth of stroke patients develop it within 2 weeks (Gamble
et al. 2000).
Lance (1996) described the complaint of a painful, burning, red ear in a CP
patient with a right sylvian infarction (F42, case 10). Some 6 weeks later she
developed sharp pains like a hot needle in the left side of her head, which recurred
with increasing frequency until it became a diffuse burning ache in the left side of
her head and face, similar to the pain she experienced in her left shoulder and
upper limb. When the burning pain exacerbated, onlookers commented that her
left ear became red and might stay red all day. Sensory loss and weakness of her
left arm persisted. Her pain was diminished to about one half of the previous
severity by imipramine 125 mg daily. Three years after the accident, she developed
left-sided migraine-like headaches associated with increased intensity of her burning pain.
Although there are no formal studies of the interaction of mood and pain state
in CPSP, the experience of chronic pain can lead to depression, anxiety and sleep
disturbance. As such, inquiries as to the length and quality of sleep as well as the
patients mood should be made.

SPECIAL CONSIDERATIONS
1. Sensory epilepsy
Epileptic pain is rare (0.3
2.8%: review of Scholz et al. 1999; 4.1%: Nair et al.
2001), although the exact frequency remains to be determined (amounting in the
United States to no less than 15 000 patients, assuming an average pain prevalence
of about 1%). Painful auras have been recognized as such for a long time
(De Ajuriaguerra 1937). Ferres textbook also reports instances of atrocious tearing
pains during jacksonian fits, although De Ajuriaguerra noted that painful fits
appeared to be less frequent than implied by Ferre, and this was in fact the general
impression of the time. Rather, it was not clear whether the origin was thalamic or
cortical (Garcin 1937). However, Penfield and Gage (1933) described the case of
an epileptic woman in whom seizures were heralded by a sharp pain in the right
lower quadrant of the abdomen, immediately followed by loss of consciousness.
At operation, they found atrophy of a small convolution just posterior to SI
and near the midline. Galvanic stimulation of this area reproduced her pain and
this was confirmed in another patient (case 5) with postraumatic epilepsy and
a normal cortex. They observed that seizures beginning in the postcentral gyrus
may be initiated by pain and discomfort in the opposite side of the body and
without direct reference to the thalamus. Other cases too of pain and paresthesias
in the same distribution had a march implying contralateral SI involvement

Central Pain of Brain Origin: Epidemiology and Clinical Features

(e.g., Young and Blume 1983). One epileptic patient had pain reproduced by
neurosurgical stimulation of parietal BA5 (see in Scholz et al. 1999).
In the partial review of Scholz and colleagues (1999) of the literature, pain
generally accompanied simple partial attacks, with or without a jacksonian march,
and with no side prevalence, in both adults and children. Pain could involve
the whole hemisoma (or sometimes the whole body or limbs bilaterally), a limb or
hemiface, combinations of these, or also spread contralaterally; also described
were visceral pain and throbbing, pricking or diffuse headache. The usual cause was
a tumor (meningiomas, gliomas, metastases or abscesses) or rarely penetrating
head trauma and stroke. In several cases, it was idiopathic.
Actually, during the attack, the patient may complain of unpleasant sensations

numbness, pins-and-needles, intensely unpleasant but difficult to define, burning,
cramping, aching, gnawing, throbbing, stinging, electric shock-like, stabbing, like
a thousand bee-stings, like a sharp knife
besides true pain; these anomalous
sensations are like those described by CP patients. In recent series, the parietal region
(SI and SII) was the commonest
with exceptions
site for lesions responsible
for the painful seizures; however, the site of the lesion may not always correlate
with the site of the seizure during ictal pain, especially if the pain does not occur early
in the ictal sequence. Bilateral EEG anomalies during painful fits are on record
(Scholz et al. 1999).
Importantly, there is no objective sensory deficit (e.g., Retif et al. 1967); instead, it
seems clear how the decreased inhibition accompanying a seizure interferes with pain
control mechanisms in certain cortical areas. This might account for the apparent
intensification of paresthetic or dysesthetic sensations to the point of becoming
painful, in some patients.
2. Parkinsons disease (PD)
Pain as part of PD was recognized by Parkinson himself (Garcin 1937). According to
Garcin, PD-associated pains
siegent principalement aux membres, a` la nuque et aux lombes, occupant surtout les
articulations et les muscles sous forme de douleurs profondes parfois atroces ou
survenant par crises delancements et de brulures, surtout nocturnes. Elles sont souvent
limitees au cote atteint dans les syndromes unilateraux. Tre`s souvent, ces douleurs
prece`dent les debuts apparents de la maladie . . . Il est plus rare de les voir persister tout
le long de la maladie.
(principally affect the limbs, the nape and the loins, mostly at the level of joints and
muscles as deep, sometimes atrocious, pains or shooting or burning painful paroxysms,
mostly at night. They are often limited to the affected side in unilateral syndromes. Very
often, these pains precede the onset of the disease . . . more rarely they persist
indefinitely.)

A current estimate is that one third to one half of patients suffer some form
of pain. Sage (2004) classifies these into low-DOPA pain states (dystonic,
pseudoradiculopathic, akathesic, genital, trigeminal neuralgia-like, abdominal,
nonspecific, musculoskeletal and paresthetic, generally described as burning, tingling,
numbness in distal limbs or groins), high-DOPA pain states (dystonic, choreic,

101

102

Central Pain Syndrome

paresthetic, which is generally described as burning in limbs or trunk) and others

(oral, vaginal, joint, muscle tightness, headache, gastrointestinal, hemifacial dystonic,
depression-associated).
Forster (1927) believed that the striopallidal system exerts an inhibitory action
on the thalamus, possibly explaining CP apparently due to striatal lesions. However,
stereotactic lesions of the globus pallidus for the treatment of extrapyramidal
motor disorders never originated CP. Honey and colleagues (1999) improved one
of two PD patients with poorly localized, bilateral, often burning dysesthesias
with pallidotomy at 6 weeks postoperatively, but none at one year; instead, cramping
and deep aching pains responded to pallidotomy, with most patients relieved
or improved at 1 year. In a PET study of normal volunteers, Hagelberg and colleagues
(2002) concluded that D2 receptor binding potential in the human, particularly
in the striatum, may determine the individual cold pain response and the potential
for central pain modulation: an individual with only few available D2 receptors
in the forebrain is likely to have a high tonic level of pain suppression, combined
with a low capacity to recruit more (dopaminergic) central pain inhibition by
noxious conditioning stimulation. Hodge and King (1976) found that, following
induction of sensory loss in humans, L-dopa increases pain and the area of
denervation, while methyldopa does the opposite.
However, the central nature of at least some PD-associated pains has been called
into question by Djaldetti and colleagues (2004). These authors found that: (1) PD
patients (n 36) have significantly lower heat pain thresholds than matched controls
(while tactile and warm thresholds did not differ), (2) patients with painful PD have
significantly lower heat pain thresholds than pain-free PD, (3) heat pain threshold
is lowermost in the most affected limb and (4) there is no difference between ON
and OFF phases. This study definitely rules out PD pain as a CP.
As regards dystonia, despite suggestions that part of the pain may be centrally
mediated, the nature of the dysfunction is unknown and we feel not enough human
evidence is available to advance the argument beyond speculation.
3. Iatrogenic lesions (Table 2.13)
Several neurosurgical operations can originate CP. These are briefly reviewed.
Hemispherectomy. In cases of infantile hemiplegia, CP is generally short-lived
(weeks), being more persisting at long-term follow-up in cases operated on for
cerebral tumors.
Mesencephalotomy. Unlike bulbar and spinal tractotomies, these also damage the
epicritic pathways (medial lemnisci and tracts of Goll and Burdach); moreover, they
invariably impinge on the midbrain reticular formation both in open and stereotactic operations. At midbrain level, the spinothalamic tract consists of only a small
number of fibers (about 1500; Glees and Bailey 1951), since a large share stopped in
the reticular formation, and the collaterals of the fibers severed at mesencephalic
level may still convey pain impulses on the polysynaptic system of the brainstem.
Mesencephalotomy also impinges on descending inhibitory systems centered on the
periacqueductal gray.
Bulbar tractotomy. The low incidence of CP after bulbar tractotomies has been
explained by the fact that the surgical incision interrupts both spinoreticular and

103

Cortectomy of the posterior lip of Rolandos

fissure for amputation stump pain

n
y

Riechert (1961)

Cassinari et al. (1964)

Destruction with radioactive yttrium (2 pts)

Talairach et al. (1960)

4. Destruction of thalamoparietal radiations

Tasker (1990)

Lewin and Phillips (1952)

3. Parietal cortectomy

Petit-Dutaillis et al. (1950)

2. Frontal lobotomies, topectomies, lobectomies

13 cases

Zuelch (1960) (see also

Mueller et al. 1991)

Quarti and Terzian (1954)

y

Laine and Gros (1956)

Gardner et al. (1955)

y
y

Obrador (1956)

6 cases

CP

Brain tumors (2 pts)

Type of lesion

Bell and Karnosh (1949)

Dandy (1933)

1. Hemispherectomy*

Author(s)

(continued)

Delayed (3 mos) in 1 brachial plexus avulsion pain, CP in regions not affected by

the original pain

Review: no CP ever followed parietal cortectomies

Transient (2 mos) trigeminal paresthesias and overreaction to pain and thermal

stimuli

In the majority, transient (1 week in one)

Dysesthesias, thermal allodynia (long term)

Transient hyperesthesia in the limbs in all (2 weeks)

Transient hyperpathia and allodynia (movement) up to 1 week

As above

Transient (bar a trace) hyperpathia; allodynia

Allodynia in both

Notes

TABLE 2.13. Iatrogenic lesions originating central pain (selected studies)

104

Type of lesion

y
y

Vc, DM

Bilat CM

Vc, CM (1 pt, 90Y); Vc-CM (2 pts); Vc-medial

lemniscus at subthalamic level-CM (1 pt);
CM (1 pt); CM first and Vc later (1 pt)

Obrador et al. (1961)

Urabe and Tsubokawa

(1965)
Cassinari and Pagni (1969)

n
y
y

Nashold (1974)

Pagni (1977)

Tasker (1990)

y
y

Dogliotti (1938)

Walker (1942a)

Open lateral tractotomies

7. Mesencephalotomies{

Cassinari and Pagni (1969)

y?

Vc

Arcuate, CM, DM

Bettag and Yoshida (1960)

CP

Mark et al. (1960)

Vc

Hassler and Riechert (1959)

6. Thalamotomies

Pagni (1977)

5. Lesions of lenticular and caudate nuclei and adjacent areas

Author(s)

TABLE 2.13 (continued)

2/2 a few days to a few weeks later

Almost immediate in 2/4 pts (another died)

Review; stereotactic lesions of VL, VA, posterior lateral, dorsal median, centrum
medianum, anterior, intralaminar, pulvinar and reticular nuclei sparing Vc and its
basal part never cause CP
Review; no CP after lesions confined exclusively to the medialis dorsalis, ventralis
lateralis, anterior thalamic nuclei or pulvinar
Review; CP follows thalamic surgery damaging or destroying Vc, when complete
anesthesia is not obtained, immediately after lesioning
CP follows always in a mild degree all stereotactic thalamotomies he performed; it
never occurs after destructive lesions of the Vim and Vor nuclei to relieve
motor disorders

6/32 pts

2/7 pts

Dysesthesia in 1 phantom pain pt

In 3/7 facial painful anesthesia pts (recurrence of the original pain or true CP?)

Painful dysesthesias in 1/77 pts numbness in 2

Delayed CP in 1/24 pts (phantom pain)

Review: no CP ever from pure lesions

Notes

105

Bilateral lateral mesencephalotomy

y
y

Nashold (1982)

Frank and colleagues

(1982)
Tasker and Dostrovsky
(1989)

6/52 cases

(continued)

Review of 92 reports; 15
20% of pts submitted to mesencephalotomy develop

dysesthesia

Review; 70% after open mesencephalotomy, 5% after stereotactic

mesencephalotomy
2/14 cancer pain pts

86 cases

8/54; severe in 2 cases and transient in 1; mild transient sensory disturbances

possible after ML lesions

1 PNP case; 8 mos later, dynamic tactile allodynia in hemisoma

1 of 2 (hyperpathia to painful stimuli and vibration allodynia); 1 pt died 14 days

after the operation without CP

Review; lateral mesencephalotomies have the highest incidence of CP among all

pain-relieving operations

1 week later in 1 tabetic case; all over the body

3
14 days later in all 6 pts (entirely new sensations), touch sensibility only
mildly affected
Hyperalgesia to pricking and touch allodynia in 1 cancer pt; no lemniscal deficits

2/2

Review

Mazars and colleagues

(1960)
Voris and Whisler (1975)

Lesion of the STT/QTT, a large part of

the reticular substance at the level of the
posterior colliculi
Mesencephalic coagulations of the STT

Wycis and Spegel (1962)

1. Destruction of STT/ML (complete)

reticular formation brachium inferior
colic (partial)
2. Lesion of STT/ML extending to the most
medial portion of the thalamus

Roeder and Orthner (1961)

Torvik (1959)

Stereotactic mesencephalotomy

y?

Lateral mesencephalotomy

Drake and McKenzie

(1953)
Bailey and colleagues
(1954)
Mikula et al. (1959)

Cassinari and Pagni (1969)

y
y

Lateral mesencephalotomy

Schwarz (1950)

Walker (1950)

106

Type of lesion

1 bilateral

Tractotomy of the descending

trigeminal root

White (1962)

Never immediate; 465% of cases in 32 pts (many with cancer and tic), 2/35
unbearable, 18/35 milder; in some pts transient paresthesias. Paresthesias may
increase or decrease with time
Unpleasant paresthesias
y

2 pts; hyperpathia and allodynia

20 pts; disagreeable, but not bothering, sensations

Zuelch and Schmid

(1953)
White and Sweet
(1955)

12/28 pts, very severe

2/6 pts

descending
obex level
descending
obex level

2/4 pts (both neuropathic pains)

Review of 63 mostly cancer pain cases; bulbar spinothalamic tractotomy only

rarely gives rise to CP. This may depend on the surgical interruption of
spinoreticular fibers intermingled with STT fibers, before reaching bulbar reticular
formation; deep incisions injure reticular neurons. No injury to lemniscal fibers
ensues with bulbar and spinal lesions

1/8 pts

1 pt (burning hyperpathia to deep pressure)

Notes

Hamby and colleagues

(1948)
Falconer (1949)

Tractotomy of the
trigeminal root at
Tractotomy of the
trigeminal root at

CP

Grant (1948)

Le Beau et al. (1948)

Olivecrona (1947)

9. Bulbar trigeminal tractotomy (Sjoqvists operation)

Cassinari and Pagni

(1969)

Zuelch and Schmid

(1953)
Crawford and Knighton
(1953)
Crawford (1960)

8. Bulbar tractotomy

Author(s)

TABLE 2.13 (continued)

107

Thoracic

Zuelch and Schmid

(1953)

White and Sweet

(1955)
Horrax and
Lang (1957)

Cervical plus thoracic

Falconer (1953)

Lapresle and Guiot

(1953)

5/58 pts unpleasant sensation of heat or cold; 6/58 pts burning pain

y
y

Sjoqvist (1950)

Sasaki (1938)

Miserocchi (1951)

y
y

Dogliotti (1937)

(continued)

Icy cold sensations (in 1, 419 mos), paresthesias more severe after cervical
cordotomies (43 yrs); deep aching and shooting pains
2/50 pts

Hyperalgesia plus hyperpathia; tactile allodynia in 4/6 phantom pains;

continuous sensation of pins-and-needles in 1/6 phantom pain pts;
burning pain
Hyperpathia; tactile allodynia, hyperalgesia

5/8 arthrosis pts; hyperpathia; at least in one case new pain exactly comparable
to thalamic pain

Heat allodynia? Troublesome paresthesias

Also in complete pain paths transection

7/19 pts; postoperative paresthesia

A few cases out of 47

Burning or dysesthetic sensations

Burning or dysesthetic sensations

Review

Review; never immediately, 1 week or later; 30% of cases, with wide range; usually
paresthesia, rarely pain, generally not as intense as after Fraziers surgery

Babtchine (1936)

Open technique

12. Cordotomies

Jefferson (1983)

Botterell et al. (1954)

11. Cordectomies

Pagni (1977)

10. Pontine lesions

Cassinari and Pagni

(1969)

108
y

y
y

Bohm (1960)

Brihaye and Retif (1961)

Diemath et al. (1961)

20%

y
y

Ganz and Mullen

(1977)
Kuhner (1981)

Cowie and Hitchcock

(1982)
Nathan and Smith
(1972, 1984)

High percentage of dysesthesia, with 1% of worrying burning dysesthesia after 1 yr

Mazars (1976)

Dysesthesia: 3/49 after unilateral cordotomy; 1/7 after bilateral cordotomy

Dysesthesia: 1/41 after bilateral cordotomy; 3/79 after unilateral cordotomy

y
y

3%

Cancer pain; average survival: 6 mos

y
n

Mansuy et al. (1976)

Incidence: about 1%

3 severe and 9 mild dysesthesias out of 276 cancer pts. Delayed or very delayed.
More likely to affect pts who lived longer (allowing analgesia to give way to
hypalgesia); 3 severe and 6 mild dysesthesias out of 50 upper thoracic
cordotomies and 1 severe and 3 mild dysesthesias out of 30 C1
2 cordotomies in
non-cancer pts
19% incidence in non-cancer-related pain. Generally long-lasting, commonly
persisting until death of pt. Unlikely to be relieved by a new cordotomy

4/35 unilateral cervical cordotomies; 4/34 bilateral thoracic cordotomies

(generally paresthesias-dysesthesias); in 1 pt a further higher cordotomy did not
suppress burning paresthesia
1/109 pts; intense burning sensations in the analgesic
territory 3 weeks later
1/121 pts (dysesthesia)

Dysesthesias in 8/105 C7
D3 incisions and in 4/98 C2
C5 incisions. Burning

sensations below the level of analgesia; in only 4 hypalgesic zones involved.
Spontaneous improvement in most of the cases; pain persisted over 10 years in
one pt
Dysesthesias in 2/120 pts

Notes

Rosomoff (1969)

Percutaneous technique

White and Sweet (1969)

Schwartz (1960)

CP
y

Type of lesion

Grant and Wood (1958)

Author(s)

TABLE 2.13 (continued)

109

Incision at the C5
6 level impinging

on lamina II and Lissauers tract
Pure dorsal column incision

y
y

Sourek (1969)

Broager (1974)

Lesions of nc cuneatus

Sjoqvists operation

Surgical interruption of the posterior

columns in 2 pts with intramedullary tumor
Bilateral section of the tracts of Goll

Wertheimer and Lecuire

(1953)

14. Commissural myelotomy

Pagni (1977)

White and Sweet (1955)

Browder and Gallagher

(1948)

Grant and Weinberger

(1941)
Pool (1946)

Antonucci (1938)

Forster (1927)

13. Posterior cordotomy

La Huerta et al.
(1994)
Tasker and North
(1997)
y

Lipton (1989)

2 of 34 until death 3
6 mos later

Most of 24 pts, transiently

(continued)

Girdle pains in 27/107 pts; dysesthesias-paresthesias in 28/107 pts (plus

hyperalgesia). May be persistent

Review. No case among 43 pts

Numbness in one leg, one-night pain; complete disappearance of symptoms after

about 3 weeks
Paresthesias attributed to the extension of the lesion to the ipsilateral cuneate
nucleus
Allodynia; thermal hyperalgesia in large girdle area of torso (from incision level down
to lower 3
10 dermatomes)
Intermittent burning type of pain in the stump of the arm (?)

Hyperpathia and spontaneous pain likely due to other spinal tracts damage

Severe dysesthesias in <2% of percutaneous cordotomies; higher incidence of not

troublesome dysesthesias
6% painful dysesthesias at/below lesion level in 181 cervical anterolateral
cordotomies (developing, in one case, 200 days later)
3.1% at discharge in 244 pts, persisting in 1.6%; burning or dysesthesias after a
delay in all or part of the body rendered hypalgesic or analgesic
Review: data in many series difficult to analyze

110
y
y

Sindou et al. (2001)

Falci et al. (2002)

Sometimes perception of touch or pin-prick as dysesthesia or unpleasant

sensation, but not pain. In the early period after surgery some pts developed
dysesthesia
4.7% (VAS 1
3)

Dysesthesias

CP does occur in some pts but the incidence is not high

Mild tactile allodynia in 1/5 in analgesic areas

In 9 pts, immediately after surgery, occasional dysesthesias, prominent or mild; no

severe dysesthesias or hyperpathia in the long term

Temporary paresthesias or dysesthesias in most pts, disappearing within a few

days; long-lasting in 4/24 pts
12 of 24 pts developing immediate transient CP (max. 2 weeks)

Notes

* In cases of infantile hemiplegia, symptoms are generally short-lived (weeks); in tumoral resections, they last longer. Thalamus may be injured during surgery.
{
At midbrain level, the STT consists of only 1500 fibers, since 90% stopped in the reticular formation in the brainstem; mesencephalic lesions may impinge on descending analgesia stations: this may
relate to enhanced responsiveness to painful stimuli. Midbrain lesions invariably impinge on reticular formation.
{
At times, after STT-tomies, a vivid girdle pain appears, on one or both sides, usually referred to the transitional area between the analgesic and normal skin on the side contralateral to the incision and
accompanied by hyperesthesia; usually disappears in a few weeks; may be due to trauma to the exposed spinal roots.
NB: no CP after lesion of motor or extrapyramidal fibers.

Powers et al. (1984)

Nashold (1984)

16. Dorsal root entry zone coagulation (DREZ)

Hyndman (1942)

15. Section of the tract of Lissauer

Cook and Kawakami

(1977)
King (1977)

CP
y

Type of lesion

Lippert et al. (1974)

Author(s)

TABLE 2.13 (continued)

Central Pain of Brain Origin: Epidemiology and Clinical Features

spinothalamic fibers, before the former reach the nucleus gigantocellularis in the
bulbar reticular formation. Also the reticular formation may be injured if the incision
is too deep. No injury is normally caused to the lemniscal fibers.
Bulbar trigeminal tractotomy (Sjoqvists operation). Trigeminal pain-relaying fibers
give off collaterals which end in the other portions of the trigeminal complex sensory
stations and in the reticular formation, and these are not involved in Sjoqvists
operation. A polysynaptic intranuclear pathway, similar to the supposed ascending
polysynaptic pathway of spinal lamina II, transmits impulses from the caudal nucleus
to the rostral portion and thence to the reticular formation (Stewart et al. 1964).
Thus, if the bulbar trigeminal tractotomy interrupts only the descending tract of the
trigeminus and not the nucleus with its intranuclear pathway, extratrigeminal
impulses may still be transmitted, giving rise to sensory dysesthesias, while
interruption of the nucleus too should block impulses ascending in the intranuclear
pathway, making the occurrence of dysesthesias more unlikely.
Anterolateral cordotomy (open and percutaneous). The incidence ranges from 0 to
more than 90% (Mazars 1976), being higher in patients undergoing cordotomy for
the treatment of benign disease (e.g., 4% in cancer pain and 19% in non-cancer pain
in White and Sweet 1969); this is accounted for by the latent period necessary for CP
to arise. Symptoms are generally long-lasting, commonly persisting until the patients
death, relief by a new cordotomy being unlikely. A few without any postoperative CP
are on record (e.g., Mansuy et al. 1976): this might also depend on short follow-up
time (6 months). Mazars (1976) pointed out that a higher incidence in some series
may depend on more extensive damage to the cord during surgery. Recent series can
be as low as 1
3% or as high as 20% (reviewed in Tasker and North 1997).
Appearing a few days to many months after a successful operation, these are
generally feelings of icy cold, burning dysesthesias or pain, sometimes with hyperpathia and hyperesthesia, generally referred to areas in which pain sensibility is
recovering, but also to totally analgesic areas. They are often most pronounced in
the original painful area for which the cordotomy was performed. CP can be
distinguished from a relapse of the original pain: in the latter, the pinprick sensation deficit disappears in the painful area due to a retraction in the pinprick level,
while in CP the pinprick level is retained and pinprick sensation is absent in the
painful area. At times, following anterolateral cordotomies, a vivid girdle pain
appears, which radiates to one or both sides, usually referred to the transitional area
between the analgesic and normal skin on the side contralateral to the incision and
accompanied by hyperesthesia. According to Sweet and Poletti (1989), girdle pain
disappears usually in a few weeks, being due to temporary trauma to the exposed
spinal roots.
Rarely, following anterolateral cordotomies, both mono- and bilateral (but also
after vascular damage to the anterolateral quadrant), the patient perceives pain and
temperature (but also non-painful) stimuli applied to analgesic or hypalgesic regions
in a part of the affected or controlateral side of the body in which the sensibility is
normal (referred or reference of pain when the pain is felt in a place apart from the
spot where the noxious stimulation is applied; mirror pain, allochiria, allachesthesia
when patients misperceive the location of a stimulus at the same point on the
opposite side of the body), a phenomenon first described by Obersteiner (1881).

111

112

Central Pain Syndrome

In cases of unilateral cordotomy, pain is usually referred to the symmetrical contralateral part of the body; in cases of bilateral anterolateral cordotomies (or vascular
damage), giving rise to bilateral analgesia, it is referred to the ipsi or contralateral
side above the analgesic zone. The patient reports that the pain slowly spreads, as
stimulation is maintained, and arises from the interior, unlike the stimulus to the
skin which is felt as external (Nagaro et al. 1993). However, referred pain is not CP,
as a further cordotomy on the opposite side abolishes it (Chapter 8).

CENTRAL PAIN OF CORD ORIGIN

Central pain of cord origin is also known as below-level pain, remote pain,
functionally limiting dysesthetic pain syndrome. Burning dysesthetic pain and central
dysesthesia syndrome are general terms that have been used to describe CCP too.
1. Lesions causing CCP (Tables 3.1 and 3.2)
CP has been reported with virtually every type of disease or lesion affecting the spinal
cord substance (dorsal horns), be it a complete or an incomplete lesion. Trauma/
concussion (civilian gunshot wounds and automobile accidents in western countries)
is the leading cause of CCP worlwide, but iatrogenic lesions dismayingly follow suit.
CP, although only one of the many chronic pains observed after SCI (Table 3.3), is
by far the most severe and disabling, and in many patients may limit their functional
ability and daily activities.
2. Incidence and prevalence (Table 3.4)
Literature series are not comparable, because pain terms used are not homogeneous
and research methods vary widely (e.g. subjective self-reports versus objective study);
moreover, CCP can be simulated by other concurrent pains, making it difficult to
tease out, and in most series there is no agreement on what true CP is. Thus,
quoted estimates of CCP in the literature range from a few to almost all. Burke
(1973) even reported different incidences of pain among paraplegics in different
societies, which he blamed on some aspects of patient management. Prospective
studies with enough power have not been published, but Siddal and colleagues (1999)
found that almost 20% of their SCI patients developed below-level pain at 6 months
and Bonica (1991), in reviewing a total of 2465 SCI patients in the literature, found
that no less than 25% had CP. CP is next in order of frequency among SCI pains after
end-zone pain; however, in the 41-patient series (36 ASIA A) of Falci et al. (2002),
below-level pain was the predominant pain, occurring in 31 (end-zone in 8,
simultaneously in 2). In the United States and the EU, there are about 600 000 SCI
patients and 150 000 may be suffering CP; worldwide, 2.5 million spinal cord injured
patients are estimated to exist.
Injuries that result in severe damage or disruption of the spinal cord and
its adjacent tissues (e.g., gunshot wounds) as well as those with large intraspinal
113

114

Central Pain Syndrome

TABLE 3.1. Causes of cord central pain (compiled from a complete survey of the

literature and personal observations)

(1)

Spinal trauma with fracture and/or dislocations producing complete or partial transection or
concussion of the spinal cord

(2)

Ischemic/hemorrhagic (e.g., aortic dissection, systemic hypotension, atherosclerosis/thromboembolism/infarcts, hematomyelia*/subarachnoid hemorrhage due to AVMs{, cavernomas, dural fistula,
traumatic/nontraumatic/iatrogenic cervical anterior spinal cord syndrome, etc.)

(3)

Rheumatological and degenerative disorders (e.g., myelopathy due to cervical spinal stenosisspondylosis and cervical discal hernia, ankylosing spondylitis with conus lesions, Pagets disease,
rheumatoid arthritis, posterior longitudinal ligament ossification)

(4)

Intra- and extramedullary tumors{

(5)

Congenital and developmental (nontumoral cysts, syringomyelia, dysraphism, diastematomyelia,

spina bifida, myelomeningocele, etc.)

(6)

Inflammatory-infective (multiple sclerosis: transverse myelitis, viral (e.g., herpes zoster, cytomegalovirus, HIV, poliovirus), bacterial (e.g., mycobacteria/Potts disease, luetic gumma}), fungal (e.g.,
cryptococcus) or parasitic infections/abscesses (e.g., toxoplasma, schistosoma) or infective
transverse myelitis)

(7)

Degenerative CNS disorders

(8)

Toxic (antiblastic agents, radiation, etc.)

(9)

Genetic and metabolic

(10) Iatrogenic (cordotomy, aortic repair surgery, surgery for spinal angiomas/fistulas/hernias/
spondylosis/intra- and extramedullary tumors, spinal fusion surgery, myelography, anticoagulant
therapy with epidural/subdural hematomas)
* Sudden at-level pain, sometimes followed by below-level pain.
{
Initially produce at-level pain, then commonly below-level pain.
{
Cervical-thoracic extramedullary tumors generally produce long-lasting at-level pain and shorter-lasting below-level
pain more often involving the lower limbs. Pain or dysesthesias can be the only (or initial) symptom for a long time.
Intramedullary tumors generate less frequent, below-level (short-lived) pain/(long-lived) dysesthesias, often in both
legs and at-level (armor-like constrictive band).
}
The pathological process in tabes dorsalis, which can originate CP, is known not to be confined to the posterior columns
(Vierck 1973).

TABLE 3.2. Distribution of causes of CCP

Series A* (%)

Series B{ (%)

Trauma
Tumors

65
6

75.3
6.2

Inflammatory
Infective
Skeletal
Vascular/ischemic
Congenital (or uncertain [A])
Iatrogenic

5
3.6
2.5
1.2
1.2
10

2
2
4
12

(gunshot
closed trauma)

(ependymoma, meningioma,
schwannoma, etc.)
(MS, etc.)
(cervical stenosis, etc.)
(syrinx, etc.)
(surgery for cervical disk (2.5% in B),
radiotherapy others, etc.)

* Tasker et al. (1992) (127 CCP patients seen between 1961 and 1989), Canada.
{
Rogano et al. (2003) (81 patients seen prospectively), Brazil.

Central Pain of Cord Origin

TABLE 3.3. IASP classification of SCI pains (adapted from Siddall et al. 2001)
Nociceptive
(1) Musculoskeletal (mechanical and lesional pain). Dull, aching, movement-related, eased by rest,
responsive to opioids and NSAIDs. Located in musculoskeletal structures. Bone, joint, disk, ligament,
muscle and soft tissue trauma and inflammation (e.g., strain in latissimus dorsi in a C7 complete
quadriplegic); mechanical instability, muscle spasms, secondary overuse syndromes. It may add to and
compound certain end-zone pains. Muscle pain is caused by stress consequent to mechanical
deformity due to immobility or overuse of shoulders, arms and back, when innervated, for balance and
mobility purposes, but also to secondary changes following fractures and fixation, mechanical
instability, and osteoporosis. Muscle spasms may sometimes produce discomfort and cramping pain in
the legs and abdominal muscles. Pain is referred at injury level or right above it (including spasmrelated). Especially when acute, it usually recedes with treatment.
(2) Visceral pain. Dull, cramping. Located in abdominal region with preserved innervation. Renal calculus,
bowel dysfunction, sphincter dysfunction, etc. Also includes dysreflexic headache. It usually presents in
high-thoracic and cervical SCI (quadriplegic) patients, despite varying degrees of sensory anesthesia
and/or paralysis, as chronic cramping pain or discomfort/fullness centered mostly in the periumbilical/
hypogastric and pelvic areas. It is both spontaneous and provoked by a full urinary bladder, fecal
impaction of the colon and rectum and other conditions that distend hollow viscera. It may be
associated with nausea, flushing of the face, headache, piloerection and sweating. It may present in the
absence of any abdominal organ dysfunction. Visceral sensation is conveyed via the dorsal columns.
Some visceral pain can actually be CP.
Neuropathic pain
(1) Above-level pain (in the region of sensory preservation). Compressive mononeuropathies (carpal tunnel
syndrome) present in up to half of paraplegics, often due to wheelchair propelling; complex regional
pain syndrome observed as arm pain and swelling in quadriplegics and rarely in incomplete
paraplegics, mostly bilaterally.
(2) Below-level pain. CCP (see text).
(3) At-level pain (also known as transitional zone, radicular/root, girdle, segmental, end-zone, junctional,
boundary zone pain). Occurs at or just above the level of the sensory loss, in the cutaneous transition
zone from the area of analgesia to areas of normal sensation (i.e., hypesthetic) and extends for 1
2
dermatomes and often more caudad (5
6 dermatomes) into the anesthetic zone; often it is not strictly
dermatomal (radiculometameric), can be uni- or bilateral (asymmetrically more than not), and can be
observed at all levels, perhaps with some preponderance, often in clinically complete injuries. It is
generally described as dull, aching (sometimes burning) with superimposed paroxysms of throbbing,
stabbing, electric shock-like or cramping pain lasting from one to several minutes. Allodynia and
hyperalgesia are frequent: touching/stroking the skin in the painful dermatomes, which may also
present as a very narrow band of hyperalgesia, often activates the pain itself, causing it to radiate into
the lower parts of the body, especially the legs. Trigger spots can also be found on the surface of the
skin, in the hypesthetic, painful areas, but sometimes as far as 6 dermatomes above the level of spinal
trauma. When these spots are touched, the pain is aggravated. Visceral stimulation (e.g., full bladder)
can also trigger it. When pain due to T8
L2 vertebral lesion is referred to the legs, many patients
present with muscular spasms of ana- or hypoesthetic paraplegic legs, and pain is spasm-related.
At-level pain is usually due to direct injury to the dorsal roots at or near the site of trauma, but
also Lissauers tract and posterior horns, or even local arachnoiditis/scarring with entrapment
(occasional worsening by arm/leg movement suggests traction on these roots). One-third of SCI
patients have it, making it the most common type of pain in association with paraplegia (Nashold
1991; Beric 1999). A subset of these pains is cauda equina pain (damage from T12 caudad), and
involves the legs, feet, perineum, genitals, buttocks and rectum. It is generally very severe; usually
burning, it may often be seen with dysesthesias and neuralgic pain in the thighs, calves or feet.
Double lesion syndrome (DLS), seen more frequently in patients with complete cervical or upper
(continued)

115

116

Central Pain Syndrome

Table 3.3 (continued)
thoracic lesions (perhaps 20%), is essentially a cauda equina/root dysfunction that modifies leg spasticity
and bladder behavior due to upper cord damage: the pain is most often sharp, pricking and electric shocklike, with occasional burning, as well as aching and dull. Although usually stabilizing within several months,
or several years, it remains constant, with minimal fluctuation. Most patients eventually adapt to it. It
generally asymmetrically involves the leg, groin, thigh or foot, often in association with the perineum, rectum
and genitals (Beric 1999). It is not CP. L5
S1 avulsion injuries of the conus (with myelocele at the L5/S1
foramen) are typically due to severe pelvic fracture (which may also cause pelvic plexus injuries). The pain
is typically confined to the leg and associated with varying degrees of weakness. Progressive posttraumatic
myelomalacic myelopathy too may lead to transitional zone pain.
Despite past claims, there is no evidence for psychogenic pain in the setting of spinal cord injury.

hemorrhages are more prone to produce pain than a compression lesion produced
by simple fracture dislocation (Nashold 1991; Tasker 2001). CCP after trauma
at levels higher than T10 has historically been considered rarer and of lesser
intensity in almost all series (e.g., Davis and Martin 1947; Freeman and Heimburger
1947), as conocaudal injury adds a peripheral component due to nerve and/or
nerve root damage (Nashold 1991). For instance, Davis and Martin reported
very severe CP in 8/77 cervical lesions, 73/288 thoracic lesions and 45/106 lumbar
lesions; previous series indicating the contrary (e.g., Holmes 1919) were written off
as small in size and with short follow-up. However, a review of data clearly shows
that CCP is equally represented at cervical, thoracic and lumbar levels and with
similar intensity; what appears to differ is the frequency of superimposed paroxysms,
higher in conocaudal injury. Also, quadriplegics may suffer more pain than
paraplegics. Neither vertebral level nor completeness of lesion affect the incidence of
steady CCP, although steady (usually burning) perineal pain occurs more frequently
with complete lesions; intermittent pain occurs equally in complete and incomplete
lesions at all spinal levels, but most frequently with lesions at T10
L2 level (57%)
(Tasker et al. 1992). Since intermittent pain is the most painful component of CCP,
this may go some way to explaining the reported lower frequency of CCP at
cervicodorsal levels.
3. Age of onset and sex distribution
Patients with traumatic CCP are generally males (about 75%) and under the age
of 40 (about 60%), reflecting younger males susceptibility to trauma. No data are
available for other lesions.
4. Time to pain onset
Similar to BCP, CCP can also start immediately or even years after insult, although
sometimes it may be difficult to ascertain it amidst several other pains. In the
consecutive series of Rogano and colleagues (2003) of 81 CCP (64.2% incomplete,
35.8% complete) patients, 43.2% of the patients developed CCP within the first week,
21% at 1
4 weeks and 35.8% after 4 weeks (mean: 110.4 weeks): thus, onset is within
a month in almost two-thirds. In the series of Tasker and colleagues (1992),
traumatic CCP was delayed in about 80% of cases, in two-thirds within 1 year of

Central Pain of Cord Origin

117

TABLE 3.4. Incidence and prevalence

No. of
patients

Author

Pathology

Beric et al.
(1988)
Milhorat et al.
(1996)

Spinal cord injury

243

Syringomyelia

137

Stormer et al.
(1997)

Spinal cord injury

901

Siddall et al.
(1999)

Spinal cord injury

100

Finnerup et al.
(2001)

Spinal cord injury

330

Patients with CP (%)

Chronic SCI patients. CP in 13/243
patients (5.3%)
Retrospective review.
Segmental dysesthesia (burning pain,
hyperesthesia, pins and needles
sensations and throphic changes):
51/137 patients (37%); burning
pain: 43 (31.4%); hyperesthesia:
41 (29.9%); pins and needles:
37 (27.0%); stretching or pressure
of skin: 17 (12.4%)
Multicenter study. Pain and/or
dysesthesia in 591 patients (66%).
Pain alone in 50% of patients;
painful dysesthesia in 11%;
distressing dysesthesia without
pain in 5%.
Below-lesion pain: 278 patients (47%)
Transitional and/or below-lesion
pain: 508 patients (86%)
Prospective longitudinal study.
Prevalence of neuropathic pain
at 6 months: at-level pain, 36%;
below-level pain, 19%
Postal survey in a community-based
sample of SCI patients. 330/436
responses. Pain or unpleasant
sensations in 254 (77%). Belowand/or at-level pain/dysesthesia
in 221 (67%)

trauma. The SCI patients of Falci and colleagues (2002) (T10
L1: 35; T4
T9: 6)
experienced pain immediately in 63% of cases, within 2 months in 84%, within 6
months in 95% and within 1 year of injury in all cases. In contrast, one-third of the
patients of Nashold (1991) developed it up to 6 years later. About one-third of
patients with a delay of up to 1 year and more than half with a delay of more than 1
year harbored a posttraumatic syrinx in the series of Tasker and colleagues (1992). In
these cases, the syrinx rather than the original injury seems responsible for the pain.
Thus, late onset of pain (and always facial pain) must raise suspicions of a syrinx.
Like CPSP, CCP usually appears with some functional recovery in more severe cases
(Beric 1999).
5. Level of lesion
In two representative series (A: Tasker et al. 1992; B: Rogano et al. 2003) for a total
of 208 patients, CCP was caused by cervical lesions in 42% (A) and 28.4% (B) of the

118

Central Pain Syndrome

cases, thoracic in 21% (A: down to T9) and 44.4% (B: up to T11?) and conocaudal in
37% (A: T10
L2) and 27.2% (B). In sum, conocaudal lesions are not the most
frequent lesions causing CCP, and that is also our experience (Canavero and
Bonicalzi 2004a) and that of others (see Beric 1999).

6. Distribution of pain
Pain may involve the entire body region below the level of injury (diffuse pain), but
usually is more intense in the sacral dermatomes, buttocks and genitalia, and the feet
(Friedman and Nashold 1986), never following a dermatomal distribution. Pain is
usually diffusely and symmetrically (although not at all times during follow-up)
referred to the parts of the body whose sensation is affected by the cord lesion;
however, a quarter complains of localized pain within a much larger area of sensory
alteration, some having a pain sharply localized to a small body part, usually the
saddle area. Tasker and colleagues (1992) found, in patients with complete lesions,
that steady pain occurred as a band at the upper level of cord damage in about 7% of
cases, diffusely below that level in less than 20%, patchily below the level in about
60% and in the perineum in 15%; in those with incomplete lesions, the pain occurred
diffusely below the level of cord dysfunction in two-thirds of cases, patchily in threequarters and as a band at the upper level in less than 20%. Patients with facial pain
(about 4%) all had incomplete lesions and a syrinx. Intermittent pain tended to run
around the trunk at the level of the cord lesion in complete cases, and shoot up and
down the body and/or the legs in incomplete lesions. While pain generally starts from
the level of injury and caudad, there may be a free area from the zone of injury to the
area of dysesthesias. In a series, the most common locations included the legs (84%),
posterior trunk (63%), anterior trunk (42%) and arms (16%; 100% in quadriplegics)
(see Beric 1999).
The bizarre distribution of CCP is demonstrated by Jefferson (1983), who broke
down his paraplegia pain patients into three groups:
1)

2)

Six patients had an area of pain on the front of, or just above, the knees
(a blob of about the same size as, or marginally bigger than, the patella),
symmetrically or with side prevalence. Invariably there were also pains
occupying the front of the thighs or else the front of the shins. One had pain
on the tops of his feet and some (very localized) pain on the back of his calves
(the only patient with a significant proportion of the pain occupying the
posterior aspect of the leg in the first two groups). Only one patient
complained of pain involving the pelvis (rectum and vagina).
Three patients described pain occupying the anterior aspect of the thighs. In
two of them the pain was symmetrical and there was no pain felt in any other
part of the body. In the third patient, the pain occupied a large part of the
front of the right thigh, extending upwards almost to the groin and
downwards to the middle of the patella. There was less severe pain in a similar
distribution on the left, together with an area of pronounced hyperesthesia in
the skin overlying the medial aspect of the left knee. Additionally, there was
slight pain behind the right knee.

Central Pain of Cord Origin

3)

The third group (6 patients) had fairly widespread pain, extending from the
groin to the feet. Unlike groups (1) and (2), the pain spread downwards from
iliac crests or groins and in half the patients it also involved the backs of the
legs. Two patients had diffuse pain down the fronts of the thighs, knees and
shins and in one of them the pain extended round the hips symmetrically into
the lateral part of the buttocks. Three of the patients felt pain as extensively on
the back of the legs as on the front, in two with involvement of the feet. One
of these patients additionally described an episodic sensation which was
likened to an explosion in the rectum. One patient had leg pains and pain
involving the lower abdomen, the genitalia and the buttocks. Two patients
with no involvement of mid-thighs, knees or shins had pains in areas that
would be covered by bathing trunks (i.e., top of thighs, lower abdomen,
buttocks plus anus and rectum either on the anterior or the posterior aspect).
One of these patients also had isolated pains around the heels and ankles.
Of the three patients who had lesions involving the D10 vertebra the pains
were distributed either throughout the leg or legs or else in a bathing trunks
distribution.

Like BCP, CCP can be felt superficially or
perhaps more frequently
deeply.
In Brown-Sequards syndrome (hemisection of the cord), on the lesion side, intense
pain spreading to the paralyzed, but not analgesic, limbs may be felt suddenly at the
moment of injury, fading away in a few days or weeks: this is not CP (Garcin 1937;
Riddoch 1938). Below-level CP is observed in the contralateral hemisoma with
respect to the lesion (end-zone pain is observed ipsilateral to hemisection).
In some cases, pain is felt contralaterally after stimulation of the affected
hypoesthetic areas (allochiria).
7. Quality
There are different pains present in different patients and also different pains present
in the same patient at different times or simultaneously. Sometimes, characteristics
change as they appear or disappear. Like BCP, there is no one quality prevailing in all
studies and CCP may be described with many terms by patients. However, the steady
component may be more often burning, but also aching, cutting, piercing, radiating,
tight, stinging, compressive or distractive; it may be dysesthetic (generally tingling
but also cold). Intermittent pain is generally described as shooting or coming in
electric shocks. Aching pain may prevail at neck
shoulder
back levels, especially in
tetraplegics, and burning elsewhere (Widerstrom-Noga et al. 2001). In the series of
Falci and colleagues (2002) (41 patients; T10
L1: 35; T4
T9: 6; at-level pain: 30.9%;
below-level pain: 69.1%), the pain was most frequently described as burning (91%)
or sharp/stabbing (61%), but also as cramping/pressure (38%), stinging/pins and
needles (23%), electrical/shooting (12%), aching (12%), cold/freezing (2%),
vibrating (2%). If pain occurred at and below level, the pain was different in
character. In the series of Rogano and colleagues (2003) (81 patients, complete SCI in
35.8%, incomplete in 64.2%), pain was burning in 86%, shock-like in 39%,
throbbing in 14.8%, pricking in 13.5% and aching in 11.2%.
In the series of Garcia-Larrea and colleagues (2003) of 32 SC incomplete
injury cases (no midline pain or complete injuries) (proximal to DRG) (MS 9,

119

120

Central Pain Syndrome

trauma 7, tumor 5, syrinx 5, spondylotic myelopathy 4, zoster myelitis 2), 22

had spontaneous continuous burning pain (68.7%), 17 crushing pain (53.1%) and
19 paresthesias (59.3%); 5 had spontaneous intermittent/paroxysmal cramping pain
(15.6%) or 17 electric discharge-like pains (53.1%). Seven also had mechanical
pain. Davidoff and Roth (1991) had 19 SCI patients; pain qualities were cutting
(63%), burning (58%), piercing (47%), radiating (47%), tight (37%), cruel (37%),
nagging (37%). SCI and syringomyelia pain may have a prominent dysesthetic
element, e.g., pins and needles and stretching or pressure of the skin. Dysesthesias
may be particularly common in incomplete spinal lesions (Davidoff et al. 1987b;
Beric et al. 1988).
8. Intensity
The intensity of the pain varies from mild, unpleasant tingling to one of the most
agonizing torments known to humans. When more components of pain are present,
the intermittent will be the more severe. The steady component is generally
fluctuating during the day and from day to day, also in bursts of activity and
cyclically (namely, every other day or even every other week) (Falci et al. 2002) and is
not always so harassing as to induce the patient to ask for medical help. Pain may be
more intense in the legs (Widerstrom-Noga et al. 2001). Generally speaking, CCP is
always very intense. In the series of Garcia-Larrea and colleagues (2003) of 32 SC
incomplete injury cases, CP was never scored less than 7. In the series of Rogano and
colleagues (2003), mean VAS score was 9.4, with pain more severe with gunshot
injuries (p < 0.001). A higher level of education may be reflected in more perceived
pain. SCI CP may or may not be perceived as worse than motor deficits
(Nepomuceno et al. 1979; Davidoff et al. 1987).
9. Components
CCP consists of three components (Tasker et al. 1992): a steady, spontaneous pain
(almost all), an intermittent, spontaneous pain (about one-third, singly found in 1%
of patients) and evoked pain (about one-half, singly in 3%). So, for instance, a single
patient may complain of episodic lightning pains down a leg, superimposed on a
continuous background of burning pain. Intermittent pain is particularly common in
patients with T10
L1 injuries, whether complete or incomplete (57%), and often
shooting down one or both legs: 69% of Taskers CCP patients with intermittent pain
had thoracolumbar lesions. The steady, intermittent and evoked components are
often associated in a single patient. The type of pain has no rapport with the causative
lesion (Tasker et al. 1992).
10. Evoked pains
Evoked pain does not depend on the vertebral level nor on the completeness of the
spinal lesion and exclusively occurs in areas of incompletely or clinically undetectable
sensory loss or as a band at the upper margin of complete sensory loss; it can be
elicited throughout the entire area of hypoestesia or only in part of it, by one or
several modalities of sensory stimulation (Tasker et al. 1992). Trigger points can be
identified even distant from areas of sensory deficit. In rare instances, evoked pain
affects skin with clinically normal sensation (hyperesthesia). Overall, evoked pains

Central Pain of Cord Origin

TABLE 3.5. Clinical features (representative series; Defrin et al. 2001)
15 SCI pts with below-level pain
Quality of pain:
burning: 73%
electric shock-like: 53%
pressing: 27%
cutting: 20%
shooting: 18%
Pain was described as deep by 93% of pts (superficial by 7%)
Pain localization (tested areas):
tight: 55%
shin: 83%
foot: 100%
Onset of pain: from several days to 8 years after the injury (within 1 mo: 46%; within 2 mos: 59%; within
6 mos: 93%).
Pain increased with the years in 73% of pts, ameliorated in 6% and remained unchanged in 20%.
Factors affecting pain sensation:
external or internal factors exacerbated pain in 94% of pts:
environmental temperature change: 70%
illness (fever, infection): 50%
changes in the fullness of GI or urinary system: 27%
mental state: 26%
Ameliorating factors in 30% of pts:
warming the room or limb: 61%
evacuation of the bladder or stomach: 46%
sport activity or work: 30%
alcohol consumption: 23%
posture change: 15%
medication (CBZ, clonazepam, baclofen and dypirone): 84%

may be less frequent in SCI than in other CPs (Beric 1999). Tasker and colleagues
(1992) observed evoked pain in about 40% of patients with complete lesions and in
about 50% of those with incomplete lesions. The Danish group (Finnerup et al.
2003a) found that 60% of their patients had allodynia (30% tactile, 26% cold, 14%
warm), while Garcia-Larrea and colleagues (2003) observed it in 7/32 CCP cases
(21.5%); the SCI patients of Falci and colleagues (2002) rarely showed allodynia (7%
touch allodynia).
CCP can be worsened by the same factors as BCP, as well as skin sores and
infections. Factors such as secondary gain or drug-seeking behavior will significantly
affect the severity and chronicity of the pain.
Table 3.5 gives summary data in a representative series.
11. Somatosensory deficits (Table 3.6)
In CCP, like BCP, temperature and pain sensation is uniformly absent or impaired,
unlike touch and vibration sensation (Beric 1999). All patients have involvement of
the STT, with very few exceptions, although STT damage can be present without CCP
(Beric 1999; Eide et al. 1996; Defrin et al. 2001; Finnerup et al. 2003).

121

122

Central Pain Syndrome

Central Pain of Cord Origin

123

124

Central Pain Syndrome

Central Pain of Cord Origin

125

126

Central Pain Syndrome

Central Pain of Cord Origin

127

128

Central Pain Syndrome

In a minority of patients, the initially documented sensory loss may fade, even
though pain subsequently develops. Some studies failed to demonstrate any
differences in STT and lemniscal function between patients with CP and pain-free
patients or between painful and nonpainful denervated areas (Eide et al. 1996;
Finnerup et al. 2003). Tasker and colleagues (1992) described a patient with traumarelated CCP, who had a normal appreciation of temperature and pinprick. Yet, the
patient was not studied with electrophysiological methods.
12. Sympathetic and other signs and symptoms
Similar comments as made for BCP apply to CCP. Sometimes (420%), following
total spinal cord transection, after the phase of spinal shock, the patient complains of
phantom sensations referred to the legs, and these are very similar to amputees
sensations, being painful, uncomfortable and unpleasant, but not disabling. They
appear early, almost immediately after SCI and vanish soon after SCI (rarely they
linger on for months) (e.g., Davis and Martin 1947; see Beric 1999). Unlike amputees,
telescoping or shrinkage of the involved body parts occurs only rarely in paraplegics
and the length or posture of the phantom do not change; in addition, they are less
vivid. Paraplegics describe sensations projected from the surface, but few postural
sensations, with both voluntary and involuntary movements of the phantoms.
Phantom sensations must be distinguished from phantom pain. CP appears when
phantom sensations fade.
13. Course
Although in some cases it lasts only a few months, if paraplegia pain persists for
longer than 6
8 months after the injury (the majority), it will become a long-term
problem. Unlike BCP which usually tends not to change significantly, except in
degree, over time, CCP may change markedly, even dramatically, over the years:
it may increase in severity for several years and even change in distribution
and quality, sometimes dramatically. The patients of Davis and Martin (1947)
complained of hot burning suddenly turning into streams of fire or pressure of
a knife being burned in the tissue, twisted around rapidly and finally withdrawn.
Some patients follow an aggressive course with intensity escalation, a few having an
abatement of pain after a few years which becomes nondisabling (Beric 1999; Tasker
2001a). In Davis and Martins (1947) series, 40 of 217 patients still experienced pain
in the long term.

Central Pain of Cord Origin

SPECIAL CONSIDERATIONS
1. Syringomyelia
Syringomyelia (in the spinal cord) and syringobulbia (in the lower brainstem) are
rare diseases. The lesion is a cystic cavity filled with CSF-like fluid, varying from a
small lesion in the dorsal part of the spinal cord over a couple of segments to huge
cavities extending from the most caudal part of the cord into the medulla oblongata
(Milhorat et al. 1996 and references therein). The largest cavities leave only a thin
layer of spinal cord tissue undamaged at the maximally cavitated regions; gray matter
necrosis and wallerian degeneration are usually seen. Cavities are thought to arise in
the center of the cord, which is where STT fibers cross the midline to reach their
position in the ventrolaterally located STT. A lesion with this location will affect the
sensibility to temperature and pain, i.e., a dissociated sensory loss will appear.
Syrinxes may be associated with Chiari I malformation, cervical disk disease/
spondylosis, basilar impression and communicating hydrocephalus; they may also be
posttraumatic and be caused by spinal cord hematomas.
Spontaneous pain and subjective sensory disturbances may often precede by many
years any other sign of this slowly progressing disease. The complaint is especially
common with posttraumatic syringomyelia. Pain is generally segmental, involving
one arm (seldom both), neck, shoulder and hemithorax, i.e., in the distribution of
the suspended dissociated sensory loss (Garcin 1968). Facial pain is frequently
reported with syringobulbia. Below-level pain (e.g., leg, generally singly) is rare, and
can be observed when the syrinx enlarges. Segmental pain is attributed to lesions of
the dorsal horns or of spinothalamic fibers crossing the midline. Sensations range
from an unbearable dysesthetic crawling to acute, severe, violent, lancinating (but
also burning, aching and pressing) pain; warmth and cold are often felt as painful,
and radiation may be present (Riddoch 1938). At first they are unilateral and
intermittent, occurring in attacks; later they become bilateral and continuous (likely
due to skewed unilateral encroachment with greater pain-inducing damage), and
may persist even when analgesia and thermanesthesia in the affected dermatomes are
complete. One series (Ducreux et al. 2006) showed that 27 out of 31 patients suffered
CP in the arm, 12 with additional pain in the neck or in the thorax, and another 5 in
both the thorax and the leg. CP extended over 2
10 dermatomes, unilaterally in 24.
Spontaneous pain occurred on its own in 11 and was associated with evoked pain in
20 (allodynia to brush 12, heat 5, and cold 11). Pain was described as burning in 23,
deep (pressure, squeezing) in 14, paroxysmal (electric shocks, stabbing) in 19;
paresthesias and/or dysesthesias (tingling, pins-and-needles) were reported in 24.
Painful dysesthesias (burning pain, pins-and-needles, stretching-pressure of the
skin, in most cases hyperesthesias) occur in 40% of syringomyelia patients (see
references in Milhorat et al. 1996) and may be more frequent in females (39 versus 12
males in the series of Milhorat et al. 1996). In that series, MRI demonstrated
extension of the syrinx into the dorsolateral quadrant of the spinal cord on the same
side and at the level of pain in 43/51 (84%) cases. In 42/51 cases, the dermatomal
pattern of pain overlapped with a segment of analgesia
anesthesia. Obvious trophic
changes were seen in 15/51 patients (29%).

129

130

Central Pain Syndrome

Quantitative sensory testing (QST) shows that all patients with syrinx have
abnormal temperature and pain sensibility, mostly pronounced with total loss of
temperature sensibility. Patients in advanced stages also have impairment of
lemniscal sensibility. For instance, Attal and colleagues (2004) performed QST before
and after surgery (3 and 9 months) in patients with cervical and dorsolumbar
syrinxes, most suffering pain. Thermoalgesic, but not lemniscal, deficits were found
in all. Spontaneous pain was generally located within an area of thermal defict, but its
intensity was not correlated with the magnitude of the deficit. Surgery induced a
significant decrease of the deficits (tactile more than thermal), but those on pain were
variable and not correlated with the effects on thermal sensibility.
Nashold (1991), in his series of paraplegics with pain, found a spinal cyst in 60% of
the patients, generally extending from the site of the spinal trauma rostrally,
involving multiple segments of the normal spinal cord. In a few patients, at
operation, two separate cysts that extended above and below the site of the trauma
were found, but they were not interconnected.
Paraplegics who suffer from a traumatic syringomyelia often develop pain
extending above injury level, even many years after injury (15 in one of the patients of
Durward et al. 1982), probably due to the slow enlargement of the spinal cyst and the
subsequent pressure on the normal spinal cord above the level of the trauma; up to
two-thirds of paraplegics with pain of delayed onset exhibit a syringomyelia. The pain
is generally sharp or aching, electrical and burning in character, being often located in
the dermatomes adjacent to the injury level, but may expand to involve higher
dermatomes. The paraplegic is often aware that his or her sensory level has risen, and,
if a spinal cyst encroaches on the cervical spinal cord, motor deficits can occur in the
arms. This pain may be activated along with diffuse visceral pain by infections of the
urinary tract or by constipation. Continuous escalation in pain is the natural course.
Shunting is generally ineffective in reversing the pain in a significant number of cases
(Dworkin and Staats 1985; Milhorat et al. 1996; Kramer and Levine 1997).
Type I Chiari malformations can originate a central cord syndrome with symptoms including pain (frequently burning), often diffuse, but also restricted to
a few dermatomes, most often in the cervical region and arms, plus dissociated and
posterior column sensory loss, due to tonsillar herniation or the associated syrinx
(Meadows et al. 2001; Bejjani and Cockerham 2001).
2. Multiple sclerosis (MS)
CP as a symptom of MS has been recognized since the nineteenth century (De
Ajuraguerra 1937). Plaques of demyelination are most frequently found in the spinal
cord, particularly in the dorsal columns, in the brainstem and periventricularly in the
forebrain. Yet, despite the difficulty in determining the exact location of the lesions
that result in CP, due to widespread dissemination in the CNS, nonetheless, the
topographical distribution of the symptoms and signs in MS appear to indicate that
many, perhaps the majority, of the MS lesions that cause CP are spinal.
The largest (1672 patients from 26 centers) and best conducted study to date found
a prevalence of CP of 18.1% (Solaro et al. 2004); trigeminal neuralgia, found in 2%,
and Lhermittes sign (9%) are also considered part of the CP spectrum. All previous
studies showed major flaws, making them unreliable or not sufficient to evaluate the

Central Pain of Cord Origin

prevalence of CP versus nociceptive pain (Svendsen et al. 2005; see global discussion
in Solaro et al. 2004). CP is correlated with increasing age, EDSS, disease duration,
but not sex. Some patients with MS have CP for a limited period during relapses
(days to months); others have chronic CP. Since the worldwide prevalence is
estimated nearly 3 million, we expect that about 500 000 patients suffer CP.
MS CP is generally dysesthetic, but also tingling, pins-and-needles, pricking, cold
or warm. Like all CPs several other qualities, singly or in combination, may be
present, particularly burning and aching; a pressing belt-like (girdle) pain at the level
of the upper border of the lesion may also be seen.
CP, when maximal, was generally described as tingling (59%), tiring (52%), taut
(45%), burning, dull and grueling (41% each) in one study (Svendsen et al. 2005).
Intensity is often high. Pruritus is also part of the spectrum (Canavero et al. 1997).
During relapses, it can affect any part of the body, in different combinations at
different levels; in chronic stages, a great majority of those affected have pain in the
lower extremities, about one-third in the arm, and one-fifth in the trunk, partially or
totally, unilaterally (one-quarter) or bilaterally (three-quarters), hemipain being
uncommon. In MS patients it is not possible to relate the time of the onset of pain to
the time of development of the demyelinating lesions, because the latter cannot be
determined with certainty. However, some patients experience CP before other
symptoms, others complain of pain along with other symptoms and signs. It tends to
be worst at night and to affect less disabled patients. The pain tends to be constant,
but can be intermittent, deep more than superficial or both, and can radiate
(Svendsen et al. 2005).
As for other CPs, there is no correlation between pain and nonsensory signs. Not
infrequently, patients have signs of involvement of the posterior column-lemniscal
rather than spinothalamic dysfunction (Moulin et al. 1988), which can be less
prominent than, for example, in CPSP (Portenoy et al. 1988). Osterberg et al. (1994)
found decreased innocuous temperature
heat and cold (90%), pinprick (60%) and
noxious temperature (80%) (plus 82% and 72% for vibration and touch); only two
patients had no spinothalamic impairment at QST (note that there is a duplicate
paper from this group [Osterberg et al. 2005] with differing data from the same study
population). A recent study that employed QST found STT involvement in all CP
cases (Svendsen et al. 2005): all CP patients had signs of STT dysfunction with
decreased or increased sensation to pinprick and/or temperature sense at maximal
pain site. MS patients with pain had decreased sensibility to touch, temperature and
vibration versus healthy subjects, but no differences between patients with and
without pain were detected in detection thresholds evaluated by QST, although
patients with pain tended to have higher vibration and tactile detection thresholds.
Results from bedside testing and QST differed: a higher frequency of pain patients
had decreased sensation to touch, vibration/joint position and warmth compared
with pain-free subjects at bedside examination. However, even on QST, there was a
tendency of lower tactile pain threshold in the pain group and pressure pain
detection threshold was lower and cold allodynia as well as temporal summation
were more frequent in the pain group.
Contrary to some belief, allodynia is found in MS patients: generally it consists of
anomalous sensations as one passes a hand over the affected area or worsening of

131

132

Central Pain Syndrome

pain or dysesthesias during movement. CP can also develop or worsen during a rise
of temperature (exercise, sunbathing), so-called Uhthoffs sign. In one study
(Svendsen et al. 2005), aggravating factors were cold in 12 patients, warmth in 5,
same position for a long time in 11, body movement including walking in 6, physical
strain in 6, touch (clothes, etc.) in 9, tiredness in 4, stress in 2 and loud noise in 1.
Alleviating factors were physiotherapy/massage/extension in 12 cases, analgesics in
11, rest in 5, warmth in 4, cold in 4, change of position in 4 and body movements
in 3. Touch allodynia was most commonly reported in CP patients: these more often
had cold and/or mechanical allodynia than patients with musculoskeletal pain (a
statistically significant difference). The frequency of temporal summation tended to
be higher in CP patients.
In MS, facial pain is at first usually identical to tic douloureux, with plaques
involving the trigeminal root entry zone. Later, with involvement of the descending
root, pain becomes continuous and disagreeable and paresthesias appear.
In MS (as well as cervical spondylotic chronic myelopathies and extramedullary
tumors, both cervical and of the foramen magnum), an uncomfortable, not truly
painful, sensation, closely resembling that produced by an electric current, can be
elicited by the active or passive flexion of the head, and radiating from the cervical to
the coccigeal region and to the four limbs, so-called Lhermittes sign (Lhermitte et al.
1927; Garcin 1968).
3. Spinal epilepsy
Tonic-clonic fits sometimes accompanied by paroxysmal burning, lancinating or
even electric shock-like pain in the legs, glutei or pelvis, both ipsi and contralateral to
the fits, have been described for extramedullary tumors, multiple sclerosis and
transverse myelitis (McAlhany and Netsky 1955; Ekbom et al. 1968; Harrington
and Bone 1981). Nathanson (1962) reported a patient with an extramedullary
meningioma at T1 presenting with paroxysms of severe burning pain, lasting about
twenty seconds in the left buttock and leg, with stiffening of the entire limb (the thigh
slowly flexed on the hip as the leg partially extended). Pagni and Regolo (1987)
reported the case of a woman who presented tonic-clonic spasms followed by clonic
jerks in the left limb, along with pain in the glutei and in the anterior aspect of
the leg: an anterior meningioma at T10 was found. In both cases, the attacks
ceased within a few days of tumor removal. Miro and colleagues (1988) described
paroxysmal pelvic pain, occurring 1
3 times a day, as a symptom of MS. Pagni and
Canavero (1993) reported a woman with paroxysms of pelvic pain resembling tic
douloureux: the pain, which was at first itching and burning, became electric shocklike as the frequency of the attacks, which always lasted a few seconds, increased in
time. MRI disclosed a dorsal extramedullary meningioma at T6
7. Carbamazepine

and, later, surgery
abolished the attacks.
The dorsal columns are known to convey visceral nociception (Willis and
Westlund 2004). In spinal cases, focal demyelination induced by compression can
induce hyperexcitable foci in the cord fibers and these foci may both discharge
spontaneously and be triggered by mechanical distortion of the cord (Pagni and
Canavero 1993). Such fibers convey nociception, making these patients instances
of CP.

DIAGNOSING CENTRAL PAIN

CP is pain due to a CNS lesion along the spinothalamoparietal path. Thus, an

appropriate lesion must be demonstrated in such a location. At the same time, the
presence of PNP, which may mimic CP (e.g., diabetic polyneuropathy in stroke
patients), but also nociceptive musculoskeletal pains, must be excluded. A common
source of diagnostic uncertainty is that symptoms of CPSP regularly occur after a
significant passage of time from the precipitating event, calling for careful
interviewing.
CP is a somatosensory symptom. Nonsensory symptoms and signs do occur in
many patients with CP, because they are a direct consequence of the lesions, which
are seldom restricted to somatosensory structures, but these may be lacking
completely.
CP is independent of nonsensory abnormalities, namely in muscle function,
coordination, vision, hearing, vestibular functions and higher cortical functions, and
these may be present at the moment of examination or have subsided. In addition,
the degree of pain and sensory abnormalities may not be necessarily correlated with
the severity of other neurologic disabilities (Riddoch 1938; Garcin 1968; Tasker
2001). The distribution of these abnormalities will overlap or contain the perceived
location of the pain.
Mental status is usually normal and CP patients are no more depressed or anxious
than other chronic pain patients; psychiatric consultation is unnecessary. The
psychological evaluation is usually done with the Minnesota Personality Inventory
(MMPI), and other more detailed psychological tests (Hamilton, Beck, etc.) as
indicated; elevation on the scales of depression, hysteria and somatization is
important in sorting out the dysfunctional state.
Only the accompanying neurologic symptoms and signs help distinguish the
different subtypes of CP, caused by cord, brainstem or brain damage: pain and
dysesthesias have the same characteristics whatever the level or etiology. Similar
symptoms can be caused not only by diseases affecting primarily the CNS, but also
by lesions neighboring the neuraxis (e.g., extramedullary tumors) and damaging the
nervous tissue only secondarily. Sometimes pain is the presenting symptom and
remains an isolated finding for a long time, as occurs in syringomyelia, and
exceptionally other diseases (e.g., spinal cord tumors).
133

134

Central Pain Syndrome

By adhering to the following ladder, diagnosis of CP will be secured in practically

all patients. Particularly in spinal cord lesion cases, CP can be missed among other
accompanying pains or be misidentified for nociceptive pain tout court. CP appears
in many disguises and therefore requires a meticulous diagnostic workup.
STEP 1. A comprehensive bedside examination should be performed, above all
probing of somatosensory functions with cotton (touch sensation), an ice cube and
a warm vial (temperature sensation) and a pin (pain sensation). Quantitative sensory
testing (QST) may be indicated when sensory loss is not readily demonstrated,
although bedside testing and QST may not totally overlap (see a review of techniques
in Dotson 1997). The best way to get a history from a CP patient is to ask about all
possible pain qualities, rather than leaving it up to the patient, who will usually
not get past the burning dysesthesia and lancinating pains. Body distribution and
any summations or gradients should be included in the description. Painstaking
and repetitious questioning is required. Experience and subtlety are required for
evaluation. A pain drawing filled out by the patient helps delineate the distribution
of spontaneous pain (a pain diary assesses intensity fluctuations and, later, response
to therapy). Sensory testing should start in an unaffected area and compared
to testing in the affected area, moving outward until skin that responds normally
to stimuli is found (and vice versa when testing for evoked pains, in order to
minimize the patients exposure to painful stimuli). Description of pain (quality,
intensity, etc.) is usually assessed with Huskissons numerical rating scale (0: no pain;
10: worst imaginable pain) and/or visual analog scale (NRS/VAS) and the McGill
Pain Questionnaire (MPQ), a multidimensional inventory of pain, the Sternbach
Pain and Intensity Profile, the Zhung Pain and Distress Scale, and others (Elliott
et al. 2003). Magnitude of CP may be inferred indirectly by self-reports or interference with social, vocational and daily life activities. CP can also be inferred from
observable behavior, including facial grimace, and abnormal movement or posture.
Afferent function may on occasion be assessed by differential blocks implemented
by either mechanical pressure (direct nerve compression or tourniquet ischemia)
or injection of local anesthetics. Mechanical methods block fibers in order of size
(Ab, mechanical, first; Ad, cooling and first pain perception, second; C, warming
and second pain perception, last) with recovery in the reverse order. These sequences
are reversed for local anesthetic blocks. Such blocks help dissect the type of fibers
subserving evoked pains, as each recognizes different mechanisms.
STEP 2. MRI of the brain and cord is the neuroimaging technique of choice
in all patients: it should reveal a CNS lesion that is consistent with the findings
on neurologic examination. However, it cannot be relied upon exclusively in
differentiating between complete and incomplete cord lesions.
STEP 3. When there is no clear-cut lesion visible on MRI and/or QST, cutaneous
stimulation of Ad fibers with pulses from an infrared or argon (or more tissue
damaging CO2) laser, which are selectively sensitive to abnormalities along the
STT activated by noxious and thermal stimuli, will usually reveal a late potential
recorded from the vertex at a peak negative wave latency of approximately 250 ms

Diagnosing Central Pain

followed by a peak positivity of about 320 ms (LEPs). A dissociation between LEP

reduction and increased pain sensation is possible (Casey et al. 1996; Wu et al. 1999).
Ordinary somatosensory evoked potentials (SSEPs) evoked by electrical stimulation
of the median and tibial/sural nerves assess Ab dorsal column-lemniscal (touchvibration) mechanoreceptive fibers: these may impaired in CP, but, as discussed, are
not the prime mover; however, the N150-P260 component of the SSEPs may be a good
correlate of subjectively experienced pain (Treede and Bromm 1991). If LEPs are not
available, the latency of the flexion reflex (RIII reflex) in patients with CP (particularly
CCP) should be investigated: this is dependent on activation of nociceptor afferents.
Lesions in the CNS leading to decreased pain sensibility result in a delay (prolongation) of this reflex following electrical stimulation of the sural nerve (Dehen et al. 1983;
Weiller et al. 1989). Some patients will show no sign of impairment at the time of
examination; of course, this does not exclude that they had this initially.
STEP 4. In doubtful cases or in order to assess therapeutic response, SPECT/PET
may be indicated as well as pharmacological dissection with subhypnotic propofol IV
challenge (see Chapters 5 and 6). The propofol test is particularly useful in differentiating CP from (unresponsive) PNP and nociceptive pain in the cord trauma
setting, but also the classic nociceptive shoulder pain of stroke patients.
EMG, thermography and regional blocks (Kingery 1997; Bonicalzi and Canavero
1999) have no place in the diagnostic approach to CP. However, abnormal lumbar
SEPs in an incomplete quadriplegic or high paraplegic may provide clues as to a DLS,
in those who present with leg or perineal segmental pain, while cauda equina versus
conus medullaris pains can be differentiated by lumbosacral SEPs and videourodynamics (Beric 1999).
Some orienting clinical features of CP include (Garcin 1937, 1968; Pagni 1977):
a)

b)
c)

d)

Pain from injury to the posterior horn of the spinal cord and Lissauers tract
is on the same side of the lesion and corresponds to the affected or
neighboring metameres. Bilateral girdle pain is typical in cases of
intramedullary tumors or syringomyelia. CP after thoracic lesions can be
confused with DLS, which is usually sacral and lumbosacral, especially if atlevel pain is also present. Under these circumstances, the pain can mimic CP,
as it appears to cover the entire area below lesion level (Beric 1999).
Pain from injury to the anterolateral funiculus of the cord is referred to the
opposite side of the body below the lesion.
Dysesthesias from injury to the posterior column or to the nuclei of Goll and
Burdach are on the same side, below the lesion, and may be uni- or bilateral.
Usually, they are of short duration. Lhermittes sign is considered to be due to
mechanical excitation of the posterior columns: it is observed not only in MS,
but also in other myelopathies.
Pain and dysesthesia due to vascular bulbar lesion (Wallenbergs syndrome)
usually have a crossed distribution: to the face on the lesion side and to the
limbs and trunk on the contralateral side. Bulbar lesions can give rise to

135

136

Central Pain Syndrome

TABLE 4.1. Examination protocol for central pain (IASP)
DEFINITION: Pain caused by lesion or dysfunction in the central nervous system
DATE OF THIS REPORT:
HISTORICAL INFORMATION:
(1) Is pain the major or primary complaint? If not, indicate the alternative (e.g., weakness)
(2) Nature of primary neurologic disability
(a) Primary diagnosis (e.g., stroke, tumor, etc.)
(b) Location of disability (e.g., left hemiparesis)
(3) Date of onset of neurologic signs/symptoms
Date of onset of pain
(4) Description of pain
(a) Location:
Body area
preferably use pain drawing
Superficial (skin) and/or deep (muscle, viscera)
Radiation or referral
(b) Intensity (1
10 or VAS categorical scaling)
Most common intensity: at maximum; at minimum
(c) Temporal features
Steady, unchanging
Fluctuates over (minutes, hours, days, weeks)
Paroxysmal features (shooting pain, tic-like)
(d) Quality
Thermal (burning, freezing, etc.)
Mechanical (pressure, cramping, etc.)
Chemical (stinging, etc.)
(e) Factors increasing the pain (cold, emotions, etc.)
(f) Factors decreasing the pain (rest, drugs, etc.)
(5) Neurological symptoms besides pain
(a) Motor (paresis, ataxia, involuntary movements)
(b) Sensory (hypo-, hyperesthesia, paresthesia, dysesthesia, numbness, overreaction)
(c) Others (speech, visual, cognitive, mood, etc.)

EXAMINATION:
(1) Neurological disease
results of CT, MRI, SPECT, PET, CSF assays,
neurophysiological examinations, etc.
(2) Major neurologic finding (e.g., spastic paraparesis)
(3) Sensory examination
Preferably use sensory chart with the dermatomes. Indicate if modalities
listed have normal, increased or decreased threshold, and paresthesias and
dysesthesias are evoked
(a) Vibratory sense (tuning fork, biothesiometer or vibrameter)
(b) Tactile (cotton wool, hair movement
include von Frey if possible, nylon filaments)
(c) Skin direction sense, graphesthesis
(d) Kinesthesia (joint movements)
(e) Temperature (specify how tested, e.g., Thermotest apparatus)
Cold (noxious and innocuous); warm (noxious and innocuous)
(f) Pinprick
(g) Deep pain (specify how tested)
(h) Allodynia
To mechanical stimuli
To cold
To heat
(i) Hyperpathia (specify how tested)
(j) Other abnormalities like radiation, summation, prolonged after-sensation

Diagnosing Central Pain

e)
f)

g)

bilateral facial pain if the lesion impinges on the descending root of the
trigeminus of one side and on the crossed quintothalamic fibers, coming
from the other side. In pontine lesions pain has a hemiplegic distribution,
that is, also includes the face contralateral to the lesion. There have been cases
of bulbopontine lesions that obey no rule in which the pains affected the
lower limbs and one side of the face.
Pain following mesencephalopontine lesions occurs on the side of the body
contralateral to the lesion, with hemiplegic distribution.
Pain and dysesthesia due to thalamic lesions have a hemiplegic distribution
and affect the side of the body contralateral to the injured thalamus.
Extension and distribution are variable, but generally pain is referred to the
extremities and face; in some cases a peculiar cheirooral distribution of
sensory disturbances is observed (Garcin and Lapresle 1954).
As in thalamic lesions, pain due to cortical or subcortical lesions is referred to
the contralateral distal parts (face, hands and feet), that is, to the regions with
the most extensive cortical and thalamic representation. Immediate onset
limb CP with ipsilateral hemiballismus-hemichorea may be typical of an
anterior parietal artery stroke (Rossetti et al. 2003).

In sum, lesion level and site can be clinically diagnosed only in cases of posterior
horn lesion (girdle pain) and bulbar lesions (crossed sensory syndrome). Except in
certain particular cases, such as the cheirooral distribution of sensory disturbances as
in thalamic lesions, it is practically impossible, on the basis of the topography and
clinical characteristics of pain alone, to distinguish between cortical, subcortical and
thalamic lesions. Bilateral pain and dysesthesia referred to the limbs, although usually
pointing to a spinal cord lesion, may be exceptionally observed after unilateral brain
lesions (see Chapter 8).

137

DRUG THERAPY

At the beginning of the twenty-first century, a vast array of interventions is available

to help patients; unfortunately the full gamut of treatments is poorly appreciated by
medical professionals and, worse yet, pain therapists, and is ill-applied. CP remains
one of the most ill-diagnosed and ill-treated entities among chronic pain syndromes,
as proved by recent literature concerning patients submitted to, among others,
gabapentin, carbamazepine, baclofen, opioids, tramadol, behavioral therapy and
psychotherapy (Helmchen et al. 2002) or phenytoin, carbamazepine, valproate,
baclofen, fluoxetine and trazodone (Fukuhara et al. 1999), all ineffective or only
poorly effective agents.
Up to now, trial-and-error has been the norm in the treatment of CP. As months
or years go by, the typical CP patient finds no or unsatisfactory relief from the
handful of drugs the average pain therapist knows and administers. Many patients
often end up intoxicated or develop important side effects, with addiction to opioids
and benzodiazepines. Useless surgical procedures can also be attempted, usually
without lasting relief. Even moderate enduring pain after any treatment can still
be crippling and in time can relapse as the patient forgets about the previous
level of suffering.
The goal of treatment is the abolition of all pain, permanently. Here, we will
attempt to make the treatment of CP less empirical and more evidence-based.
An important caveat should be borne in mind: time is not an option. CP slowly
erodes patients will, incapacitating the vast majority, sapping their resources,
and must be treated aggressively, just like a cancer of the soul. The best results
for many patients will come from combination therapy in the very first place.
Polypharmacy, by whatever route, should be the norm, rather than not, eventually
combined with neuroaugmentative therapies.
Although many would object to prepackaged strategies for CP as a whole,
we believe otherwise: pathophysiological evidence (Chapters 7 and 8) strongly
suggests a common substratum to all CPs. In addition, pharmacologic dissection
helps guide therapy in the single patient, driving us away from tradition toward
a more scientifically based approach.

138

Drug Therapy

There are basically two tiers of therapies available for CP: pharmacologic (oral
and parenteral) and neuromodulative (electrical and chemical). Other approaches
will be discussed for the sake of completeness.

ORAL AND PARENTERAL DRUG THERAPY (Tables 5.1
5.4)

There seems little doubt that neurosurgical procedures will be replaced to a large extent by
drugs, at present unknown. (A. E. Walker, 1950)

1. General comments
Common to all studies is the short follow-up (generally weeks or months, versus an
absolute optimum of 5 years or more), rarely undertaken by independent observers,
and the small size, raising the possibility of type II errors (false negatives). To evaluate
properly response, the following indications have been suggested: for moderate pain,
a 20% reduction on a 0
10 scale is minimal improvement, 35% reduction is much
improved and 45% reduction is very much improved; for severe pain, decreases
on a numerical scale (NRS) have to be larger to obtain similar degrees of pain relief.
In other words, the change in pain intensity that is meaningful to patients increases
as the severity of their baseline pain increases (Cepeda et al. 2003; see also Mamie
et al. 2000). Instead, analgesic use as a measure of outcome is probably of poor
value, as it may be complicated by dependency and coexistent nociceptive pains.
Last, but not least, in most studies, sensory and affective effects are not analyzed
separately.
As noted above, valuable time should not be wasted trying all possible effective
drugs and the clinician should focus on those with the best chances of success, over
a defined timeline (see Table 6.8). CCP is often refractory to this tier and thermal
allodynia is considered more resistant than spontaneous and tactile allodynia.
Thus, if these fail, neuromodulation should be rapidly undertaken.
a. GABA drugs. The only such drug assessed in a formal RCT is IV propofol
(Canavero and Bonicalzi 2004), a strongly hydrophobic IV anesthetic agent structurally unrelated to other anesthetic agents. Tasker (2001, and references therein)
previously reported that IV infusions of 136 mg (mean) of sodium pentothal,
a GABAergic agonist, reduced brain CP in 73% of his patients (versus none with
15 to 18 mg of morphine). In our studies, propofol effectively controlled CP
at 0.2 mg/kg (one-tenth of the narcotic ED95 in humans), 5 times as effectively
as pentothal at equipotent doses for CP (discussed in Canavero et al. 1995).
Convergent evidence shows a specific effect of propofol for CP, but not PNP,
migraine or nociceptive pains, at the doses reported above (discussed and referenced
in Canavero et al. 1995, Canavero and Bonicalzi 2004). Unlike morphine and
lidocaine, which are effective in allaying mechanical allodynia-hyperalgesia, but
not cold allodynia-hyperalgesia (Section 5.1.e), our data suggest that, in CP, GABA
modulation can allay both. Propofol analgesia shows a clear-cut post-effect: after
several hours of infusion, analgesia can last for up to 24 hours (or more with longer
duration of infusion). Propofol modulates GABA neurotransmission in different

139

140

SCI

CPSP Carbamazepine

Leijon
and
Boivie
(1989)

Drewes
et al.
(1994)

Trazodone HCl

SCI

Davidoff
et al.
(1987)

Valproate

Amitriptyline

Drug(s)

Authors

15

18

No. of
patients

VAL: up to 2400 20

AMY: 25
(morning)
50 (evening)

CBZ: 800

150

Final daily
dose (mg)

TABLE 5.1. Controlled studies of oral drugs

Rating

Double-blind,
placebo controlled,
crossover
(2  3 wks,
2 wks
washout)

Randomized,
double-blind,
crossover
(3  4 wks,
2  1 wk
washout),
placebo
controlled

Pain relief:
McGill Pain
Questionnaire.
Present pain
(rating scale
1
5)

Daily pain
intensity:
verbal scale.
Post treatment
global ratings.
Comprehensive
psychological
rating scale

Pain relief
Randomized,
double-blind,
parallel, placebo
controlled (8 wks)

Study
design

6/20 improved
on VAL
4/20 improved
on placebo

1/15 improved
on placebo

Low-quality study
(see Wiffen et al.
Cochrane review,
2004). VAL:
stepped increase
starting at
600 mg 2  day.
Dose increased

Double dummy
(identical active
or placebo). 80%
men. Stepped
increase to final
dose of CBZ
(starting at
100 mg 2 day)
and AMY (starting
at 12.5 mg 2
day). No
follow-up.
1 drop out

5/14 improved
on CBZ
10/15 improved
on AMY

NNT: 9 (95% CI:

1.8
1)

At- or below-level
pain

Trazodone effects
did not
significantly differ
from placebo

No significant
analgesic effect
for VAL. NNT VAL:
10 (95% CI:
2.7
1)

AMY, but not CBZ,

produced a
statistically
significant
reduction of pain
vs. placebo, CBZ
only from 3rd
week. NNT CBZ:
3.4 (95% CI:
1.7
105). NNT
AMY: 1.7 (95%
CI: 1.1
3.0).
Higher plasma
levels correlated
with better
analgesia.
NB: not confirmed
in other studies

Authors
conclusions

Notes

Outcome

141

MEX: 450

SCI

CP

Chiou-Tan
et al.
(1999)

Haines and
Gaines
(1999)

Ketamine

Up to 100
PO

CIT: 10
40

CPSP Citalopram

Vestergaard
et al.
(1996)

Mexiletine

DEX: up to 81

CPSP Dextromethorphan

McQuay
et al.
(1994)

2 BCP;
3 CCP
(1 MS)

11

9/4

N of 1
randomized,
controlled

Randomized,
double-blind,
placebo
controlled,
crossover (1 wk
washout,
2  4 wks)

Randomized,
double-blind,
parallel, placebo
controlled

Randomized,
double-blind,
placebo
controlled,
crossover. Integral
n-of-1 design
(2  10 day
periods)

No dichotomous
data

0/9 improved
on DEX
0/9 improved
on placebo

Daily pain diary; No effect

VAS; Likert scale during the
unblended
run-in period in
any CP pt

Pain relief: VAS, No dichotomous

McGill Pain
data
Questionnaire

Pain relief, pain

intensity, mood,
sleep, global
rating

SCI dysesthetic
at- or below-level
pain. 15 pts
enrolled,
11 completed
the study

(continued)

Intolerable side
effects in the
whole group

No significant
analgesic effect
at this low dose

No significant
19 pts with
analgesic effect
chronic pain.
1st treatment
period: DEX
13.5 mg 3  day;
2nd treatment
period: DEX
27 mg 3  day.
No long-term
clinical benefit
SSRI
No significant
analgesic effect

according to
serum levels.
1 dropout. Blind
status not clear
(serum level
measured)

142

Finnerup
et al.
(2002)

SCI

Lamotrigine

Randomized,
double-blind,
crossover,
placebo
controlled

Randomized,
double-blind,
crossover,
placebo

Randomized,
double-blind,
placebo
controlled,
crossover
(2  8 wks,
2 wks washout)

Study
design

30

No. of
patients

LAM: up to 400 30

DEX: 100.
Administration
followed (4 h)
by intravenous
infusion of
morphine
15 mg

Heiskanen
CPSP Dextroet al. (2002)
methorphan

Final daily
dose (mg)

LAM: 200

Drug(s)

Vestergaard CPSP Lamotrigine

et al. (2001)

Authors

TABLE 5.1 (continued)

Results not
broken down
according to
pain type
Mixed population
of 20 pts with
chronic pain. DEX
or placebo given
4 h prior to an IV
morphine administration (15 mg)
At- or below-level
pain. Slow LAM
increase. 22 pts
completed the

DEX had no effect

on morphine
analgesia. 8 pts
responded to
morphine after
placebo
Categorical slight
to complete pain
relief (secondary
outcome

Pain relief:
VASpi,
MPQ, QST

Pain relief:
change in
median pain

No statistically
significant
effect of LAM
in the total

No significant
effects at
lower doses.
LAM reduced
pain score
approximately
30% (meaningful
reduction for
CPSP pts).
LAM is a
moderately
effective
treatment for
CPSP. NNT
LAM: n.a.
Stepped
increased to final
dose of LAM
(25 mg 1st

2nd wk, 50 mg
3rd
4th wk,
100 mg 5th

6th wk, 200 mg
7th
8th wk).
Median pain
score: LAM
200 mg: 5;
placebo:
7. Significant
reduction of cold
allodynia. 1 pt
withdrawn
because of LAM
adverse events.
ITT analysis
(200 mg of LAM)
12/30 improved
on LAM
3/30 improved
on placebo

Pain relief:
Likert scale.
Global pain
score. Stimulus
evoked pain.
Primary end
point: median
pain score during
the last wk of
treatment

Authors
conclusions

Notes

Outcome

Rating

143

Cardenas
et al.
(2002)

SCI

AMI: 10-125
Amitriptyline
0.5
Benztropine
besilate
(active placebo)

(84)
SCI pain:
26.
Transition zone
pain: 6

study. LAM more

effective in pts
with brush-evoked
allodynia and
wind-up like pain
(7/7 pain relief
vs. 1/14 without).
3 pts withdrawn
because of
adverse events.
ITT analysis
(200 mg LAM)
84 pts. 44 pts
AMI, 40 placebo.
Aim of the study:
whether AMI is
efficacious in
relieving chronic
pain in pts with
SCI. ITT analysis
and study
completers
analysis

measure): 10/22
on LAM, 5/22
on placebo

score from
baseline

No dichotomous
Pain relief:
average pain
data
intensity (NRS,
0
10). MPQ,
BPI, FMI, SWLS,
CHART

controlled
(1 wk baseline
period,
2  9 wks,
2 wks washout)

Randomized,
double-blind,
placebo
controlled (6 wks)

(continued)

No significant
differences
between AMI
and placebo in
pain intensity (also
with regression
analysis for
different types
of pain) or painrelated disability.
SWLS 4in placebo
group. Certain
subgroups of pts
may benefit. 18%
of pts chosen to
continue AMI, but
5% chose to
continue placebo.
No significant
difference in
AMI-placebo side
effects. No CP in
some pts.

sample. In
7/8 pts with
incomplete cord
lesions LAM was
more effective
than placebo on
at- or below-level
pain. NNT LAM
(incomplete
lesions, 50%
pain relief):
12 (2
1)

144

SCI

SCI

Tai et al.
(2002)

Harden
et al.
(2002)

Topiramate

Gabapentin

CPSP Amitriptyline
extended
release
Placebo

Drug(s)

Lampl et al.
(2002)

Authors

TABLE 5.1 (continued)

TOP: 800
(titrated over
10 wks)

GAB: 1800

19

AMI: 75

9 (5)

20

No. of
patients

Final daily
dose (mg)
Aim of the
study: to
investigate,
under
controlled
conditions, the
effectiveness
of AMI for the
prophylactic
treatment of
pts with acute
thalamic stroke
in preventing
CPSP

Rating

Parallel,
randomized,
placebo
controlled study

Pain relief: VAS

and descriptor
scale

Pain relief: NPS

Prospective,
randomized,
double-blind,
placebo
controlled, crossover (2  4 wks,
2 wks washout)

Randomized,
double-blind,
placebo
controlled (1 yr)

Study
design
All pts had
lesions in the
ventroposterior
thalamic region.
AMI was slowly
titrated from
10 to 75 mg in
extended release.
CPSP in AMI
group: 4/18 pts;
CPSP in placebo
group: 3/19 pts

Results limited by
the small sample
size and low
maximum dosage
of GAB

Primary
end point:
occurrence
of CPSP
within 1 yr

GAB placebo
among pain
descriptors with
the exception of
unpleasant
feeling

Many side effects

TOP PLA
(below 800 mg). at 800 mg
TOP 4 PLA on
descriptor scale
but not VAS scale
in final 2 wks only
at 800 mg

Notes

Outcome

Nonsignificant
trend to benefit
on unpleasant
feeling, pain
intensity, and
burning sensation
only

The placebo group

showed a pain rate
of 21% within 1 yr,
compared with 17%
in those under
prophylactic
treatment with AMI.
Prophylactic
treatment with AMI
did not show any
statistically significant beneficial
effect to prevent
the development
of CPSP

Authors
conclusions

145

Levorphanol

Methadone

CP

Morley et al. NP
(2003)

Rowbotham
et al.
(2003)

MET: 10 or
20

High strength:
mean daily
dosage: 8.9

Low strength:
mean daily
dosage: 2.7

19.
CPSP: 2.
Transverse
myelitis: 1

CPSP: 5;
SCI: 3;
MS: 4

CPSP:5;
SCI: 2;
MS: 4

23. CPSP:
LEV: 0.15 or
10; SCI: 5;
0.75 to a
MS: 8
maximum of
21 capsules/day

Randomized,
double-blind,
placebo
controlled

Randomized,
double-blind,
dose-response
(8 wks)

% reduction from
base line. (A) Low
strength: CPSP: 6;
SCI: 13; MS: 9.
(B) High strength:
CPSP: 16; SCI:
30; MS: 63

Pts who
completed the
study: 15. CPSP:
3/10 (mean pain
reduction: 20%).
SCI: 4/5 (mean
pain reduction:
22%). MS: 8/8
(% mean pain
reduction: 27%).
Low strength
(% mean pain
reduction):
CPSP: 14; SCI:
13; MS: 9. Highstrength (% mean
pain reduction):
CPSP: 23; SCI:
31; MS: 63
All pts poor
Average pain
Pain relief
intensity MET vs. responders to
(maximum
traditional
placebo (VAS).
pain intensity,
MET 10 mg: CPSP analgesic
average
regimen
1: 59 vs. 65.8;
pain intensity
and pain relief): CPSP 2: 33.4 vs.
46.4; cord: 47.8
VAS
vs. 42.8. MET 20
mg: CPSP1: 66.9
vs. 66.6; CPSP 2:
26.6 vs. 47; cord:
not tested
Pain relief: VAS
(0
10 cm).
Primary outcome:
mean pain rating

(continued)

Daily dose of
10 mg failed to
reach statistical
significance.
20 mg daily
dose resulted
in statistically
significant pain
improvement. The
analgesic effects
extended over
48 hours

Capsules intake
titrated by the pt.
7 of 10 pts with
CPSP did not
complete the
study (reasons
not described).
27% of pts
withdrew. Few
pts will benefit
from opioids. SCI
pain may have
included CP. MS
pain may have
included
dysesthetic pain

146

Plant-derived
cannabis
medicinal
extracts (CME)

MS
SCI

MS
CP

SCI

Wade et al.
(2004)

Svendsen
et al.
(2004)

Levendoglu
et al.
(2004)

Gabapentin

Dronabinol

Drug(s)

Authors

TABLE 5.1 (continued)

24

24. MS:
18; SCI: 4

No. of
patients

GAB. Maximum 20
daily dosage:
3.6 g (gradually
titrated dosage)

DRO. Maximum
daily dosage:
10

1:1 CBD:THC,
2.5
120

Final daily
dose (mg)
Consecutive
series of
double-blind,
randomized,
placebocontrolled
single-patient
crossover trials
(2 wks)
Randomized,
double-blind,
placebo
controlled,
crossover
(2  3 wks,
15
21 days;
2 wks, 19
57
days; washout)
Prospective,
randomized,
double-blind,
placebo controlled, crossover (18 wks;
4 wks medication/
placebo titration,
4 wks stable

Study
design

Pain relief: NPS, GAB reduced the

VAS (0
100), intensity and the
LQ
frequency of pain,
and improved the
quality of life.
Neuropathic pain
descriptors not
relieved: itchy,

Complete SCI at
the thoracic and
lumbar level.
Neuropathic pain
for more than
6 months.
All patients
completed the
study. Mean

DRO has a
modest analgesia
on MS CP. NNT for
50% pain relief:
3.5 (95% CI:
1.9
24.8)

CME can improve

neurogenic
symptoms
unresponsive
to standard
treatments
Self-administered
sublingual spray.
Dose titration
against symptom
relief or unwanted
effects

Pain relief
associated with
both THC and
CBD was
significantly
superior to
placebo

Patients
recorded
symptom,
well-being and
intoxication
scores on a
daily basis: VAS
observer
rating (at 2 wks)
Median
spontaneous
pain intensity
(numerical
rating scale)
in the last
week of
treatment. QST
Median
spontaneous
pain intensity
significantly lower
with dronabinol
than placebo.
Median pain relief
score: VAS 3 vs. 0

Authors
conclusions

Notes

Outcome

Rating

147

Dextromethorphan

Cannabis
extract

SCI

MS
pain

Carlsson
et al.
(2004)

Notcutt
et al.
(2004)

Wide range
of dosing

DEX: 270

16

Randomized,
placebo
controlled

Randomized,
double-blind,
placebo
controlled,
crossover
(2 separate
administrations)

maximum
tolerated dose,
4 wks crossover
medication/
placebo titration,
4 wks stable
maximum
tolerated dose
Pain relief: VAS
(0
100)

Study population:
15 pts with
neuropathic pain
of traumatic
origin. Most
patients
experienced
adverse effects,
none of which
was considered
severe
Side effects
comparable to
psychoactive
drugs

Benefit at no
fixed dose

effective dose:
2235 mg.
Dysesthetic pain
included.
Below-level pain
not specifically
mentioned

No effect in 1 pt,
69% VAS
reduction in 1 pt

sensitive, dull
and cold

(continued)

A single high
dose of DEX
has an analgesic
effect (up to
30% pain
reduction vs.
placebo) in
patients with
neuropathic
pain of
traumatic
origin

148

Whole-plant
cannabis-based
medicine (CBM;
delta-9tetrahydrocannabinol:
cannabidiol
[THC:CBD])

Rog et al.
(2005)

MS

Drug(s)

Authors

TABLE 5.1 (continued)

Study
design
Randomized,
double-blind,
placebo
controlled,
parallel group
(5 wks: 1 wk
run-in, 4 wk
treatment)

No. of
patients
66

Final daily
dose (mg)

THC 2.7
CBD 2.5
(each spray).
Gradual
self-titration
to a max.
48 spray/day

Notes
Inclusion criteria:
pts with
spontaneous
or evoked
dysesthetic pain
(burning, aching,
pricking,
stabbing, and
squeezing) and
pts with painful
tonic spasms.
CBM generally
well tolerated
(1 pt withdrawn
because of
adverse effects,
but more pts
on CBM than
PLA reported
dizziness, dry
mouth, and
somnolence)

Outcome
Trial completed
by 64 pts (97%):
32 CBM pts
(2 withdrawn),
32 placebo pts.
ITT analysis.
Results at week 4
(mean change):
pain intensity:
CBM:
2.7
(95% CI:
3.4
to
2.0). PLA:

1.4 (95% CI:

2.0 to
0.8),
p 0.005).
Sleep
disturbance: CBM:

2.5 (95% CI:

3.4 to
1.7).
PLA:
0.8
(95% CI:
1.5
to
0.1),
p 0.003)

Rating
Daily pain and
sleep
disturbance
(11-point
numerical rating
scale).
Neuropathic Pain
Scale (NPS).
Cognitive
function, mood,
MS-related
disability,
Patients Global
Impression
of Change

CBM delivered via

an oromucosal
spray, as adjunctive
analgesic
treatment. Mean
number of daily
sprays: CBM 9.6 +
6 (range 2
25)
placebo: 19.1 +
12.9 (range
1
47). THC:CBD
ratio approximately
1:1 (other
cannabis-based
compounds <10%).
CBM was superior
to placebo in
reducing the mean
intensity of pain
and sleep disturbance. PIGC: no
difference between
the proportion of pts
rating themselves
as much or very
much improved in
the CBM group (9/
34) vs. PLA group
(4/32). Cognitive
side effects: limited
to long-term
memory storage

Authors
conclusions

149

Oral opioids

Morphine IV

Portenoy et al. CP and

(1986)
others

CP and
others

Bainton et al. CPSP

(1992)

Naloxone

Morphine

Kupers et al.
(1991)

CP and
others

Morphine

Portenoy et al. CP
(1990)

Arner and
Meyerson
(1988)

Drug(s)

Authors

Controlled trial

Randomized,
single-blind,
placebo
controlled trial

Study
design

Randomized,
double-blind,
crossover,
placebo
controlled

CPSP: 4; SCI: 2 Double-blind,

placebo
controlled
crossover study

CCP (1 pt)

3 SCI, 1 syrinx,
1 spinal AVM,
2 CPSP

No. of
patients

NAL: up to 8 mg 20
in 20 mL
vehicle

0.3 mg/kg IV

15 mg

Dosage

TABLE 5.2. Controlled studies of parenteral drugs

Statistically
significant
reduction of
pain affect
rating (from
62 to 43).
Pain sensory
rating not
affected, with
trend toward
increasing
Transient pain
relief: 3/20 with
NAL, 4/20 with
saline, 4/20
with both
Pain relief:
VAS, verbal
pain scores

Ineffective

Outcome

Affective and
sensory
dimensions of
pain sensation:
101-point rating
scale

Rating

(continued)

Authors
conclusions

IV NAL
Pain scores
ineffective
obtained
on CPSP
immediately
before and after
NAL or saline
injection.
Subjective
ratings followed
for 2 wks

Some benefit

Acute boluses.
SCS and PVG
DBS: ineffective

Results not
broken down
according
to type

Notes

150
0.25 mg/kg
(IV bolus over
5 min)

6; CPSP: 2

CP

Backonja
et al.
(1994)

Ketamine

4-aminopyridine Escalating total 8

(4-AP)
dose from 18.0
to 33.5 mg
(IV, 2 separated
(2 wks)
infusions over
2 h)

SCI

No. of
patients

Hansebout
et al. (1993)

Dosage

Drug(s)

Authors

TABLE 5.2 (continued)

Randomized,
double-blind,
placebo
controlled,
crossover

Randomized,
double-blind,
crossover

Study
design
Effects
persisted up to
48 h after
infusion

Significant
temporary
neurologic
improvement,
including
reduction in
chronic pain,
in 5/6 pts with
incomplete SCI.
No effect was
detected in
2 pts with
complete and
1 severe
incomplete SCI
Pain relief in
CPSP pts.
Ketamine: pt
D: 50%
(ongoing pain);
pt F: 100% pain
relief (ongoing,
allodynia,
hyperalgesia).
Placebo: pt D:
0%, pt F:
modest
Neurological
motor and
sensory
evaluation

Pain rating
scale 0
10

Notes

Outcome

Rating

Pain relief lasting

2
3 hours. Ketamine
affected the evoked
pain and associated
after-sensation more
than ongoing constant
pain. Allodynia,
hyperalgesia and
after-sensation
improved. Side effects
during single-dose
injections mild
and well tolerated

Authors
conclusions

151

Hamamci et al. ?
(1996)

SCI

Calcitonin

1  100
IU/day (IM)

(3 infusions,
2 h apart)

26

9
Ketamine (KET) 60 mg/kg
(IV bolus,
6 mg/kg/min
for 17
20 min)
Alfentanil (ALF) 7 mg/kg (IV
bolus, 0.6
mg/kg/min for
17
20 min)

Placebo
controlled

Randomized,
double-blind,
crossover

CPSP: 8; SCI: 8 Double-blind,

placebo
controlled,
crossover

Eide et al.
(1995)

0.2 mg/kg
(single IV
bolus);
0.3 mg/kg/h
(continuous
IV infusion)

Propofol

Canavero et al. CP
(1995)

Effect lasting
no more than
20 min
(generally
10 min)

Pain score

Pain score of
the calcitonin
group was
significantly
lower than that
of the control
group

ALF
KET 4 PLA

Continuous and KET

ALF 4 PLA
evoked pain
relief

Pain relief:
VAS (0
10)

Neither
ketamine nor
alfentanil
significantly
changed
thresholds for
the sensation
of heat pain.
No clear
differential
effects on
at- and
below-level
pains
Post-stroke pts
with hemiplegia
and RSD.
4 wks study.
26 pts received
calcitonin,
16 saline

Continuous
(6
24 h)
IV infusion
in propofolresponsive pts.
Temporarily
effective with
hours-long
post-effect

(continued)

Uncertain
diagnosis.
CPSP in
some pts?

Both continuous
and evoked
pains were
markedly
reduced by
ketamine and
by alfentanil.
Bothersome
dizziness in
one patient

Pain and allodynia

abolition in propofolresponsive pts.
Propofol did not
reduce non-CP,
nor did placebo

152

SCI

Amobarbital

4
7 mg/kg
(IV infusion,
7
10 min,
max. dose
500 mg or
50 mg/kg)

DIA: 0.2 mg/kg/h

(mean dose:
52.1 mg)

Diazepam

Mailis et al.
(1997)

FEN: 5 mg/kg/h 3
(mean dose:
873 mg)

Fentanyl

No. of
patients

Dellemijn and CP
Vanneste
(1997)

Dosage

Drug(s)

Authors

TABLE 5.2 (continued)

Placebo
controlled

Randomized,
double-blind,
active placebo
controlled,
crossover (drugs
infused at a
constant rate for a
maximum of 5 h)

Study
design
Maximum relief
of pain intensity
was better with
FEN than with
DIA (66%
(95% CI
53
80) vs.
23% (12
35))
or with saline
(50% (36
63)
vs. 12%
(4
20)). FEN
CP pts
responders:
1/3. Placebo
CP pts
responders:
0/3

Outcome

Pain relief: VAS. VAS reduction

Sensory testing from about 6
to about 4.
Dramatic
reduction of
allodynia.
Substantial
reduction of
hyperalgesia

Pain relief:
rating scales
(including
unpleasantness)

Rating

17 NP pts.
1 pt with C4
myelopathy
(AVM)

Mixed
population of
53 pts with
neuropathic
pain. DIA as
active placebo.
Saline as inert
placebo.
2 consecutive
double-blind
infusions: FEN
DIA and FEN
saline

Notes

DIA had no clinically

significant effect on
pain intensity and pain
unpleasantness. The
beneficial effect of FEN
was independent of the
type of neuropathic
pain and the degree of
sedation. FEN therapy
produced equal relief
of pain intensity and
pain unpleasantness.
DIA and saline did not
reduce either pain
index. Side effects
more common with FEN
than with DIA or saline.
No severe side effects.
The clinical characteristics of neuropathic
pain do not predict
response to opioids
No benefit on
deep pain.
Sympathetic
block responder

Authors
conclusions

153

SCI

CP

Potter et al.
(1998)

Attal et al.
(2000)

Lidocaine

Fampridine-SR
(sustained
release
4-aminopyridine)
5 mg/kg
(IV infusion
over 30 min)

CPSP: 6; SCI:
10 (syrinx: 5;
SCI: 3;
spondylotic
myelopathy: 2)

26
12.5 and
17.5 mg bid
(PO, 2-wks
treatment period,
1 wk washout)

Incomplete SCI
in all pts

Post-study
follow-up:
12 pts took
oral mexiletine
(400
800 mg/
day) for 4
12
wks. 30
50%
relief in 3 pts
(2 lidocaine
responders,
1 placebo
responder). No
improvement
in 8 pts (6
lidocaine
responders).
Intolerable side
effects from
long-term mexiletine. Difference
between lidocaine and PLA:
moderate. In 7
pts refractory to
all previous
treatments,
spontaneous
pain responded
less to lidocaine

No statistically
significant
benefits on
measures
of pain
Pain relief
450%: 11/16
with lidocaine;
6/16 with PLA.
3 had no benefit
or worse pain
vs. 8 with PLA.
2 pts had more
relief with PLA.
Burning totally/
partially relieved
in 6 vs. 2 (PLA),
paresthesias
abolished in
8/11 vs. 2/11.
In 5 pts (62%),
allodynia
reduced 50%
by lidocaine
(vs. 1 by PLA),
in 4 by 100%
for up to 1 hour
postinjection
(never with PLA)

Patient
satisfaction,
sensory scores,
motor scores
Pain relief: VAS
(0
10), global
assessment,
QST

Randomized
double-blind
dose-titration
crossover
Randomized,
double-blind,
placebo

PLA (saline)
controlled,
crossover

(continued)

Significant greater
pain relief starting
15 min
postinjection and
lasting up to
30 min after the
end. With lidocaine,
significant
brush-induced
allodynia and
static mechanical
hyperalgesia
reduction. No effect
on thermal evoked
pains. In 2 pts,
30
50% relief
for 2
10 days.
NNT: 5 (SCI pts).
Side effects in
two thirds
of patients

154

Morphine

Attal et al.
(2002)

CP

Drug(s)

Authors

TABLE 5.2 (continued)

16 mg IV
(mean dosage,
range 9
30)

Dosage

Study
design
Randomized,
double-blind,
placebo
PLA
(saline)
controlled,
crossover

No. of
patients
15. SCI:
9; CPSP: 6

Outcome

Pain relief: VAS No significant

(1
100), QST difference in
pain reduction
between
morphine and
placebo. 3 pts
100% relieved
at the end of
injection (vs. 1
with PLA),
2 for 42 h,
1 pt worsened
by morphine.
1 syrinx pt with
prominent
mechanical
allodynia 100%
relieved

Rating
Morphine effect
correlated with
decreased
responses to
suprathreshold
thermal stimuli
(general
antinociceptive
activity).
Following the
completion of
the study all pts
began to take
sustained oral
morphine
(mean dosage:
93 mg; range
60
140 mg)
in a long-term
study on
efficacy and
side effects

Notes

Morphine significantly
reduced brush-induced
allodynia but had
no effect on static
mechanical and
thermal evoked
pains. Ongoing
pain was not
significantly
reduced, but 7 pts
(46%) responded
to morphine. The
effects of IV morphine
correlated with
those of oral
morphine at
1 month. Oral
morphine was
effective only in 3
(2 SCI, 1 CPSP)/
14 pts (1 lost to
follow-up) at
12
18 months
with 50
75%
relief, starting
from week 1
and peaking at
week 4. Morphine
PO less tolerated
than IV

Authors
conclusions

155

Morphine

Propofol

Kalman et al. MS
(2002)

Canavero and CP
Bonicalzi
(2004)

14

0.2 mg/kg
44. CPSP:
(single IV bolus) 23; SCI: 21

Up to 1 mg/kg
over 20 min,
continuous
IV infusion

Randomized,
double-blind,
placebo
controlled,
crossover

Single-blind,
placebo (saline)
controlled.
Followed by
naloxone

Pain relief:
VAS (0
10);
NVS (0
4)

Pain relief
(spontaneous
pain intensity
reduction
430% or
allodynia
reduction
450%):
24/44 pts with
propofol,
6/44 pts with
placebo

Pain relief: VAS 4 pts were

opioid
responders
(no pain relief
from placebo,
450% pain
reduction with
morphine and
425% pain
increase with
naloxone).
Effective dose:
43, 47, 50 and
25 mg
Study aimed
at validating
IV subhypnotic
propofol as a
diagnostic test
for CP

(continued)

Propofol was
significantly
superior to the
placebo in
reducing the
intensity of
spontaneous
ongoing pain
(for up to 1 h
after the
injection) and
of both
mechanical and
cold allodynia.
In a few cases,
only the evoked
components
were abolished

Morphine is
effective only in
a minority of pts
(29%) and only
at high doses.
Same results
reported by
these authors
in discussion
for CPSP

156

Lidocaine

5 mg/kg
(IV infusion
over 30 min)

2.5 mg/kg
(IV infusion
over 40 min)

Lidocaine

Finnerup et al. SCI

(2005)

0.4 mg/kg
(IV infusion
over 40 min)

Ketamine

SCI

Kvarrnstrom
et al. (2004)

Dosage

Drug(s)

Authors

TABLE 5.2 (continued)

Randomized,
double-blind,
placebo
controlled,
crossover

Randomized,
double-blind,
three-period,
three-treatment,
placebo
controlled,
crossover

10

24

Study
design

No. of
patients

Ketamine, but
not lidocaine,
showed a
significant
analgesic effect
in SCI-CP. Pain
relief not
associated with
altered
temperature
thresholds or
other changes
of sensory
function. Lidocaine and
particularly
ketamine were
associated with
frequent
side-effects
SCI at and
below level
pain is reduced
by IV lidocaine
irrespective of
the presence or
absence of
evoked pain.

Primary
objective of the
study: to
examine the
analgesic effect
of ketamine and
lidocaine on SCI
below-level
pain. Secondary
objective: to
assess sensory
abnormalities
to identify
responders.
Sensory
assessments
do not predict
response to
treatment
26 pts with
NP at or below
level enrolled,
2 dropped out
before any
treatment.
Evoked pain
in 12 pts.

Pain relief: VAS. Positive

QST, traditional response (50%
sensory tests
reduction in VAS
score during
infusion):
5/10 pts with
ketamine,
1/10 pts with
lidocaine,
0/10 pts with
placebo.
Temperature
thresholds:
no changes.
Sensibility:
no changes

Pain relief:
VAS. QST

Neuropathic
at-level and
below-level
spontaneous
pain:
1: significantly
reduced in all
patients

Authors
conclusions

Outcome

Notes

Rating

157

(p < 0.01)
2: significantly
reduced in
12 pts with
evoked pain
(p < 0.01)
3: significantly
reduced in
12 pts without
evoked pain
(p < 0.048)
No difference in
number of pts
with pain
reduction 33%
between the
patients with
(n 6) and
without (n 5)
evoked pain.
At-level
brush-evoked
dysesthesia
significantly
reduced.
Median pain
reduction: about
35%. NNT
for 50% pain
relief: 3
No evoked
pain in 12 pts.
Adverse effects:
IV lidocaine,
19 pts;
placebo, 1.
No correlation
between
maximal
plasma
concentration
and maximal
pain relief or
pain intensity.
No significant
decrease in
cold allodynia,
pinprick
hyperalgesia,
or pain evoked
by repetitive
pinprick

(continued)

Lidocaine is
usually not
suited for
long-term
treatment

158

Drug(s)

S()-ketamine

Authors

Vranken et al. CP
(2005)

TABLE 5.2 (continued)

50 or 75 mg
daily
(transdermal
iontophoretic
administration)

Dosage
33
(CPSP, 8; MS, 1;
PD, 1; thalamic
lesion, 3;
brainstem
lesion, 4;
SC lesion, 16)

No. of
patients
Randomized,
double-blind,
placebo
controlled

Study
design

1 week trial.
Appropriate
dose from an
open-label
preliminary
study.
Sample size
and power
calculated
pre-study
(with 33 pts,
power 0.8 for
estimated VAS
differences).
Only mild and
spontaneously
resolving
adverse events
without

No statistically
significant
differences in
VAS between
ketamine (both
dosages) and
placebo

Ketamine
50 mg:
7.3 vs 6.2
Ketamine
75 mg:

Pain intensity:
VAS.
Health status
(PDI, EQ-5D).
Quality of life
(SF-36).
Safety
assessment

Placebo group:
7.1 vs. 6.4

Pre- vs. posttreatment VAS

scores:

Notes

Outcome

Rating

Iontophoretic
administration
of S()ketamine was
no more
effective than
placebo in
reducing pain
scores, but daily
administration
of 75 mg of
S()-ketamine
improved health
status and
quality of life

Authors
conclusions

159

7.3 vs 5.7
No
improvement in
health status or
quality of life
from ketamine
50 mg.
Significant
improvement in
PDI, EQ-5D and
SF-36 (except
for role-physical
functioning and
general health
perception)
from ketamine
75 mg
differences
between
ketamine and
placebo groups.

160

161

162

163

164

165

166

167

168

169

170

171

IV 1.5 mg/min
up to 30 mg

IV 25 mg/min

Morphine

Pentobarbital

Plotkin
(1982)

IV 50
225 mg
(average: 136 mg)

IV 15
18 mg

Pentothal

Morphine (some
also fentanyl)

Tasker (1984)

IV 3 mg/kg
(infusion 240 mg)

Lidocaine

Boas et al.
(1982)

IV

Atophanyl

Di Biagio
(1959)

Route/dosage

Drug(s)

Authors

TABLE 5.4. Uncontrolled studies: parenteral drugs

BCP/CCP

BCP/CCP

CP, thalamic: 1

Thalamic pain:
1; SCI: 3

CP: 2

Pain type;
no. of patients

Case series

Case series

Placebo controlled

Study design

0%

73% responded
to therapy

90% relief (transient)

No response
to morphine
Results not broken
down according
to pain.

Great relief in
1, 0% in another

Outcome

(continued)

CP is not dependent
on opiate
mechanism; 55% of
CCP cases responded
to morphine, but only
the evoked pains and
less frequently
lancinating pains,
rarely steady pain

1.5
2 mg/kg
enough for CP

Pentobarbital administered
until the pt is on the point of
unconsciousness, at which
time pain should totally
disappear if central

Some control attempted,

partially single blind. Pain
relief evaluated by a 10 point
scale. Several days between
morphine and pentobarbital
test. Morphine followed by
naloxone 0.8 mg. Morphine
saturation test method
as per Hosobuchi.

Other details

172

Drug(s)

Lidocaine

Naloxone

Naloxone

Lidocaine

Lidocaine

Lidocaine

Lidocaine

Thyamilal, morphine

Authors

Edwards et al.
(1985)

Budd (1985)

Bowsher (1988)

Backonja and
Gombar (1992)

Edmonson et al.
(1993)

Galer et al.
(1993)

Nagaro et al.
(1995)

Migita et al.
(1995)

TABLE 5.4 (continued)

Doses as per
Yamamoto et al.
(1997)

IV. 1.5 mg/kg

in 1 min

5 mg/kg/h for
60
90 min IV

IV. Initial bolus:

50
100 mg
continuous infusion
for 48 h

IV. Single infusion

IV 0.8 mg

20
50 mg IV

IV

Route/dosage

CP: 2

CP and PNP

Case report

Case series

Retrospective series

Case series

CPSP: 4

CP: 13

Case series

Study design

BCP: 6; CCP: 2

Cordotomized pts
with mirror pain

CP

CP

Pain type;
no. of patients
Other details

Pt 1: barbiturate and
morphine ineffective.
Pt 2: barbiturate
effective, morphine
ineffective

Study on TMS

SCI pain relatively

Pain score measured by
visual analog scale (VAS, refractory
0
10), 5, 15 and
35 min after the infusion
decreased to less than
50% of pre-infusion
value in more than 75%
of cases of thalamic pain

1 excellent relief, 3 partial

reliefs, 9 0% reliefs

All patients reported some

relief within the first 12 h
of infusion. Subsequent
oral mexiletine trial: 2 pts
had excellent relief at
1 yr, 2 stopped because
of intolerable side effects

3 BCP benefited over

8
20 weeks; partial
relief in 2 SCI cases

Increased pain in
one third and induced
it in one not suffering it

Benefit

Benefit

Outcome

173

Morphine

Amobarbital

Koyama et al.
(1998)

Ketamine

IV. 50 mg

25 mg IV
CPSP: 1

CPSP: 5; CCP: 3

IV 50 mg every 5 min
up to 250 mg
IV 5 mg every 5 min CPSP: 23
up to 25 mg.

Thiamylal

CPSP: 39 (thalamic
25, extrathalamic 14,
brainstem 0)

IV 3 mg every 5 min
up to 18 mg

Morphine, then
naloxone

Kumar et al.
(1997)

Yamamoto et al.
(1997)

Case report

Case series of DBS

Case series

IV (but not PO)

AMO was effective in
reducing CP, although
similar plasma
concentration levels
were reached PO
and IV

0% relief in all

Study evaluating the

effect of IV morphine
(day 1), IV thiamylal
(day 2) and IV ketamine
(day 3). A few pts fell
asleep with barbiturate.
Threshold of significance:
40%. No differences
between thalamic and
suprathalamic cases.
All ketamine responsive
cases except 1 also
sensitive to thiamylal,
but 4 cases resistant
to ketamine responded
to thiamylal.

(continued)

CP after loss of his left upper

extremity. IV AMO was followed
by 300
400 mg/day AMO PO

1 thalamic and
1 suprathalamic pts
(of 3 sensitive to
IV morphine) relieved
at long term by
oral morphine
30
120 mg die
All pts refractory to
imipramine (75 mg),
maprotiline (60 mg),
bromazepam
(12 mg), ibuprofen
(600 mg)
Pain worsened by
ketamine in 2 pts

174
IT, 50 mg
IV, 50 mg
IV, 0.6 mg/kg
6 mg/kg/min
IT

Midazolam

Baclofen

Fentanyl

Alfentanil

Thiopental

Lidocaine

Herbal cannabis

Wajma et al.
(2000)

Trentin and
Visentin (2000)

Chatterjee et al.
(2002)

Clonidine

1 joint daily

4 mg/kg IV over
30 min

IV (approximately
1 mg/kg)

IV, 3
5 mg/kg in

30 min
IV, 60 mg/kg
6 mg/kg/min
IT, 1
2.5 mg

Lidocaine

Ketamine

IV, 0.2 mg/kg

Propofol

Canavero and
Bonicalzi 1998

Route/dosage

Drug(s)

Authors

TABLE 5.4 (continued)

CPSP: 1

CP: 16

SCI: 1

CCP: 1

Pain type;
no. of patients

Case report

Case series

Case report

Review with case

report

Study design

Later good response

to mexiletine PO, but
not amitriptyline

Original CP decreased after

16 subarachnoid blocks with
local anesthetic. IV thiopental
was the most effective
treatment in CP. CP worsened
by spinal anesthesia

All drugs proposed as

diagnostic test

Other details

Complete pain relief and Right hemiplegic

painful dystonia
marked improvement
(left-sided idiopathic
in dystonia from

IM butorphanol, saline
and atropine sulfate as
placebo, IT morphine HCL,
mexiletine, IV lidocaine
ineffective. IV thiopental,
fentanyl, butorphanol,
ketamine, midazolam,
droperidol, sevofluraneoxygen anesthesia
quite effective
44% responded; after
45 min, LIDO PLA

IT midazolam effective;
propofol effective;
baclofen and other
agents not as effective

Outcome

175

Morphine

Lidocaine

Ketamine

Morphine

Willoch et al.
(2004)

Cahana et al.
(2004)

Cohen and
DeJesus (2004)

Nuti et al.
(2005)

IV test

PCA device
(2.7 mg/h basal;
same dose on
demand)

IV, 5 mg/kg (in

150 mL of saline)
over 30 min without
a bolus. 2 daily
cycles for 5 days at
a 6 month interval

PO

Case series

Case report

CCP (syrinx): 1

CP: 7 pts

Persistent spontaneous
pain and frequency of
pain attacks reduction
was observed
immediately, 1, 3 and
7 days and 1, 2,
and 3 months after
treatment in all body
areas, but the chin

Case report

CP (postinfective
pontine lesion): 1

No significant effect

One year later, pain

dramatically decreased,
opioids significantly
reduced

Poorly effective

CPSP: 2

smoked cannabis
(3 months follow-up)

Previous high dose

opioids ineffective

Persistent pain relief after

repeated IV lidocaine
infusions. PET study.
Thalamic hypoperfusion
renormalized. CP unresponsive
to amitriptyline, nortriptyline,
carbamazepine,
oxcarbazepine, gabapentin,
valproate, lamotrigine,
baclofen and clonazepam

caudate atrophy).
3 temporarily
successful
thalamotomies
performed. Partial
response to morphine
plus buproprion and
amitriptyline (VAS
reduction from
9/10 to 4/10)

176

Central Pain Syndrome

ways from barbiturates and benzodiazepines, although IT midazolam, but not

oral benzodiazepines, reduces CP in propofol-responsive patients (Canavero and
Bonicalzi 1998, 2004). Most importantly, propofol at doses effective for CP has an
exclusive GABA A action, without appreciable effects on other transmitters/
modulators and ion channels (see complete references in Canavero and Bonicalzi
2004). PET studies reveal that propofol at subhypnotic doses first targets the cortex,
then the thalamus; it is thus possible that subhypnotic propofol allays CP by
renormalizing a specific derangement at cortical level (discussed and referenced in
Canavero and Bonicalzi 1998, 2004; see also Hofbauer et al. 2004). Propofol also
renormalizes brain deactivations seen in CP patients (see Chapter 7). Occasional
worsening might depend on drug increasing ongoing overinhibition at CNS sites.
Barbiturates can reduce CP, but their pharmacodynamic profile goes beyond
simple GABA agonism and may even reduce (like halothane, a volatile anesthetic
with a GABA A profile) descending inhibition, favoring long-term increases in
nociception (Sandkuehler 1996). Thiopental is administered IV at 50 mg boluses
up to 225 mg and thiamylal at 50 mg IV every 5 minutes up to 250 mg: when
effective, relief appears after 5
8 minutes and lasts several minutes (Migita et al.
1995; Mailis et al. 1997; Yamamoto et al. 1997; Koyama et al. 1998). There is no
experience in treating CP paroxysms with IV boluses of midazolam (3
5 mg),
but this deserves consideration. Oral benzodiazepines have no effect on CP, with the
possible exception of clonazepam on painful paroxysms. Benzodiazepines enhance
GABA A-mediated inhibition within thalamic reticular nucleus and thereby suppress
GABA B-mediated inhibition in relay neurons (Huguenard and Prince 1997).
Baclofen, a GABA-B agonist, has relieved CP via the IT route (Chapter 6), but no
meaningful analgesia is generally seen at orally tolerated doses (<60 mg die).
On the basis of these studies, we distinguish two classes of CP: GABA responsive
(Class A) and GABA refractory (Class B) (Canavero and Bonicalzi 2004). GABA
responsiveness (Class A) marks patients who stand the best chances of relief from
neuromodulation (Chapter 6). There may also be a differential of responsiveness
between disruptive and compressive CP-related lesions, with the former less
responsive than the latter (unpublished observations), but this awaits confirmation.
Several other oral drugs display GABA agonism (Moshe` 2000) and may reduce CP.
Gabapentin increases GABA levels in the human brain via a poorly characterized,
non-GABA receptor-mediated mechanism, with highest affinity for locations in
the outer layers of the frontoparietal (and other) cortices. This drug has relieved
some patients (particularly evoked pains in one series); however, in our extensive
experience, no more than 5% of CP patients are relieved in monotherapy at
2400
3600 mg and, in line with other investigators (Ness et al. 2002), gabapentin
does not have proven utility as a monotherapy in the experience of . . . (SCI)
patients. Even in a recent randomized, controlled study which also included nine
CP patients, relief greater than 50% was seen in only 21% of patients versus 14%
on placebo over 8 weeks (Serpell et al. 2002). Despite a reputation for tolerability,
actually gabapentin has an incidence of major side effects no different from carbamazepine (Wiffen et al. 2000) and in SCI pain, clomipramine seems to be more
effective
although less well tolerated
than gabapentin (55% versus 48% of
patients) (Reboiledo et al. 2002). Progabide, a GABA A agonist, awaits to be tested.

Drug Therapy

We saw no major effect with vigabatrin, an irreversible GABA transaminase

blocker, which diffusely increases GABA levels. Thus, drugs such as vigabatrin and
perhaps tiagabine (a selective GABA reuptake blocker) and pregabalin, but also
topiramate, might not be the appropriate choice, as GABA is excessively increased
unselectively throughout the brain (unlike lamotrigine and gabapentin); anesthetic
gases (GABA A agonists) also do not relieve CP. In the human brain, vigabatrin/
tiagabine increase GABA 300%, topiramate 200%, lamotrigine/valproate/gabapentin
(c. 1200 mg) 150% (Verhoeff et al. 1999); after single doses, cerebral GABA rises
70% acutely (hours) with topiramate, 48% with gabapentin and 0% with
lamotrigine; with target dose reached at 4 weeks, these increases are, respectively,
46, 25 and 25% (Kuzniecky et al. 2002). GABA is rapidly reuptaken by neurons
and glia, but with some drugs the excess cannot be eliminated. If confirmed, this
would be in line with the observation that there are only a few specific brain
regions in which augmentation of GABA function is anticonvulsant (Gale 1992):
enhanced GABA transmission can either reduce or increase brain excitability
depending on the brain region in which this happens. Moreover, only GABA
elevation associated with nerve terminals, but not with metabolic compartments,
can lead to enhanced GABA transmission. Differences among drugs may also depend
on electrophysiological factors, e.g., enhancement of the frequency of opening
rather than the open channel duration of the receptor. Spike-wave discharges are
blocked by GABA A agonists such as benzodiazepines or T-type Ca2 blockers
and are worsened by direct activation of GABA currents, for instance by
phenobarbital (also a non-T calcium, sodium and AMPA blocker), a drug ineffective
for CP. Also, GABA B activation results in more hyperpolarization than does
GABA A activation. However, the neuronal conductance increase and thus the
decrease in neuronal input resistance is much greater with GABA A than B;
accordingly, GABA A inhibits more by clamping the membrane at a subthreshold
level and thus shunting EPSPs, while GABA B inhibits more by hyperpolarizing
the membrane. Unlike GABA B (voltage subtractor), the GABA A response is much
more nonlinear (voltage multiplicator) and GABA A response is faster than B
(Shepherd 2004). Finally, we do not expect neurosteroids (e.g., ganaxolone) or
gamma hydroxybutyrate, naturally present substances in the brain with GABA
effects, to have major effects in CP.
b. Antiglutamatergic agents. The glutamate NMDA receptor subtype is widely
believed to play a pivotal role in pain transmission along with the AMPA subtype.
Although extensive animal experimentation concluded for a major role in the
mechanisms of neuropathic pain, actually, agents that antagonize the NMDA
receptor have basically no place in the long-term treatment of CP: their side effect
profile is unfavorable, particularly for ketamine, and the achieved benefit is no
greater than other better-tolerated drugs. Even patients with different chronic pains
selected for oral treatment after IV ketamine challenge (e.g., 0.4 mg/kg IM) do not
uniformly draw benefit from oral therapy (Rabben et al. 1999), e.g., 1.5
2.5 mg/kg
5
6 times a day of ketamine, and long-term oral treatment may result in
hepatotoxicity and severe psychic complications. NMDA blockade with dextrorphan/
dextrometorphan (D-isomer of methorphan, also a sigma opioid and 5HT agonist)

177

178

Central Pain Syndrome

or the modestly stronger adamantane drugs (amantadine and memantine) relieved

few patients to a major degree, without disabling side effects (see Canavero et al.
2002; Taira 1998); no experience has accrued with cycloserine. While no published
data on felbamate (another antiglutamatergic agent) exist, its potential for lethal
complications limits its clinical usage.
Poor or modest relief in most patients and the narrow therapeutic ratio of
clinically available NMDA antagonists speaks against their use (Sang 2002). However,
these drugs provided invaluable information on the neurochemistry of CP,
particularly IV ketamine (Canavero and Bonicalzi 1998). Importantly, case reports
of patients with cortical disease who failed to demonstrate analgesia to surgical
stimuli in spite of multiple doses of ketamine suggested that an intact and functioning
cortex is a prerequisite for ketamine analgesia (Drury and Clark 1970; Morgan et al.
1971; Janis and Wright 1972). NMDA receptors are especially important for
intracortical processing. In an fMRI study, subanesthetic doses of ketamine especially
reduced acute nociceptive-related brain activity in the insula and the thalamus
(Rogers et al. 2002).
c. Sodium channel blockers. Initially employed to quench abnormal activity in
neuromas and demyelinated peripheral fibers of PNP patients, the bulk of evidence
points to the efficacy of sodium channel blockers in CP (Tremont-Lukats et al.
2005). Both lidocaine and its oral congener mexiletine act in an activity-dependent
manner: as doses increase, repetitive firing is blocked before axon conduction,
providing a degree of selectivity for paroxysmal activity. Although a peripheral action
has been established at doses below those achieving conduction block, a central
action is also likely (Boas et al. 1982); the observed preferential antihyperalgesic
and antiallodynic effects of lidocaine suggest a selective central action on the
mechanisms underlying such evoked pains. In particular, they may have a specific
action on brush-evoked and mechanical allodynia, unrelated to general analgesic
effects (Attal et al. 2000). In healthy humans, a single IV bolus of lidocaine results
in sustained and constant concentrations in the cerebrospinal fluid, with a faster
plasma decay (Usubiaga et al. 1967; Tsai et al. 1998). Although the lidocaine test
may predict analgesia from mexiletine in several patients, this is not generally
indicated.
IV lidocaine has been administered at doses of 1 mg/kg (over 10 minutes) to
5 mg/kg (over 30
120 minutes) diluted in saline, sometimes via a pump. Pressure
and EKG monitoring are mandatory. Dysarthria and somnolence call for immediate
suspension and lidocaine is contraindicated with Adams
Stokes syndrome or severe
atrioventricular heart block. The most frequent minor side effect is dizziness during
infusion. Mexiletine is started at 200 mg and increased every few days to a maximum
of 1000 mg, analgesia or intolerable side effects; it must be administered on a full
stomach to reduce nausea. Other side effects include dizziness, tremor, jitters and
headache. A yearly EKG is indicated. Patients with a history of heart disease,
arrhythmias, atrioventricular heart blocks and other EKG anomalies must be
evaluated by a cardiologist before starting PO mexiletine. Lamotrigine must be
increased slowly starting from 25
50 mg to reach a maximum dose of 600
800 mg,
analgesia or intolerable side effects. While generally well tolerated, lamotrigine

Drug Therapy

very rarely triggers Stevens
Johnson syndrome, which can be fatal. Carbamazepine

is often poorly tolerated at effective dosage, with several CNS effects (e.g., ataxia).
It must be started at 100 mg bid and increased to a maximum of 1200 mg, analgesia
or intolerable side effects (an analog, oxcarbazepine, may be better tolerated);
moreover, white cell counts must be monitored frequently in the first few months
to check for possible hematologic toxicity. Valproate (also a weak T calcium blocker
and unspecific pro-GABA drug) has allayed pain in few patients. Clearly, safer drugs
are needed.
Similarly to GABA agonists, only some sodium blockers relieve CP and mexiletine
may be the most potent (Canavero and Bonicalzi 2005; see also Bowsher 1994),
although a formal trial is needed. Topiramate (also an antiglutamatergic, GABAergic
agent; but see discussion in Canavero et al. 2002), phenytoin (also a L Ca2 and
NMDA blocker and GABAergic agent) and carbamazepine (also a weak L Ca2
blocker, proserotoninergic, antiAMPA (
20%) and antiadenosine-1 agent) are the
least effective, except for paroxysmal MS/SCI pains (Wiffen et al. 2000). Phenytoin
has a longer time dependence for frequency-dependent blockade and for recovery
from it than carbamazepine; neither affect amplitude or duration of individual
action potentials, but both reduce the ability of neurons to fire high-frequency
trains of action potentials in a use-dependent manner. Lamotrigine (which is also
a powerful antiglutamatergic and modest GABAergic agent) has been found effective,
but it lost effect in all our initially published patients (Canavero and Bonicalzi 1996)
after 2
3 years (unpublished observations). Sodium block also leads to reduced
glutamate release and the differential potency of mexiletine and lamotrigine
versus carbamazepine may relate to this factor (Canavero and Bonicalzi 1996).
Riluzole, a sodium channel blocker and NMDA antagonist has proved scarcely
effective for PNP (Sang 2002), and we do not expect major analgesia in CP either.
Zonisamide, a sodium
T calcium channel blocker with some GABA, serotonergic
and dopaminergic effects, may prove helpful (see also Finnerup et al. 2002 for
further discussion on mechanisms of action of anticonvulsants). In contrast, agents
that target specific sodium channel subtypes in peripheral tissues (e.g., SNS1/PN3)
will most certainly have no impact on CP, although some believe that the
Nav1.3 subtype is upregulated in first, second and third-order neurons of the
pain pathway after contusive SCI, leading to enhanced excitability (Waxman and
Hains 2006).
From a pathophysiological standpoint, reviewed data point to hyperactivity as
a mechanism of CP (Canavero and Bonicalzi 1998; Max and Hagen 2000).
d. Antidepressants. Amitriptyline, a tricyclic, is useful for BCP, but not (or only
insignificantly) for CCP, syringomyelia CP or MS CP (see also Beric 1999), where
side effects have a larger impact. Even in BCP patients, only about half will benefit
and often only partially; worse yet, many cannot tolerate its many and sometimes
serious side effects. The NNT of amitriptyline and congeners in CP is 1.7 (McQuay
et al. 1996; Sindrup and Jensen 1999). Continuous, lancinating and thermally (but
less so mechanically) evoked pains may respond, with analgesic doses less
but not
much less, at least in our experience
than antidepressant ones. Analgesia can
appear days to 5 weeks after initiation, regardless of dose, and can increase slowly,

179

180

Central Pain Syndrome

even if plasma levels are stable: a trial of efficacy should never last less than 2 months,
bar intolerable side effects. After suspension of therapy, analgesia is lost gradually,
but slower than expected from plasma levels. Relief is seen in both depressed and
nondepressed patients and depressed patients may have their mood improved,
but not their CP. Side effects include, among others, orthostatic hypotension
(alpha block), urinary retention, memory loss, cardiac conduction abnormalities
(muscarinic block) and sedation (histamine block). Amitriptyline is usually started
at 10 mg in the elderly and 25 mg in younger patients and increased by 5
10 mg
every few days until benefit or toxicity or a maximum dose of 300 mg. Being sedating
(but less than doxepin), it can be given before sleep, until divided doses become
necessary. Tricyclics call for an EKG at onset and EKG and blood level measurements
above 100 mg die. Artificial saliva is indicated to counter mouth dryness (an
anticholinergic effect). Anticholinergic effects contraindicate tricyclics in SCI
pains, as these can trigger constipation or bladder retention and pain exacerbations.
Although some congeners may be better tolerated (e.g., nortriptyline and
clomipramine), it is a general opinion that amitriptyline is the most effective of all
antidepressants, with tertiary amines more effective than secondary amines and both
more than SSRIs (Sindrup and Jensen 1999). In fact, serotonin-specific reuptake
blockers (fluoxetine and congeners), but also atypical agents, such as trazodone and
risperidone, although better tolerated, have not proved superior to tricyclics and
actually relieved few patients to a meaningful extent, and so serotonin/norepinephrine (e.g., venlafaxine) and selective norepinephrine (e.g., reboxetine) blockers.
Interestingly, Sicuteri (1971) had reported that 4 out of 18 cases of intractable
migraine treated with parachlorophenylalanine and 2 out of 23 treated with reserpine
(both drugs are serotonin antagonists) developed spontaneous pain, hyperalgesia and
hyperpathia in the body, similar to CP. No experience has accrued with MAO
inhibitors. Why amitriptyline should be the most effective agent remains
controversial. The importance of norepinephrine and serotonin agonism singly has
not stood up to scrutiny (Jasmin et al. 2003; Canavero and Bonicalzi 2004). Yet,
the range of action of tricyclics is bewildering: besides NE and 5HT effects, they
also are alpha-1 and histamine-1 antagonists, sodium blockers, NMDA antagonists,
opioid agonists, anticholinergic, increase local levels of adenosine, potentiate
neurogenesis, regulate synaptic plasticity and enhance endocrine function (see
references in Jasmin et al. 2003).
e. Opioids, naloxone and cannabinoids. By the IV route, morphine (and
congeners) relieved none or only few patients with CP and only at very high
doses. Their place in the treatment of CP is very limited (Canavero and Bonicalzi
2003; see also Warms et al. 2002; Eisenberg et al. 2005); their long-term side effect
profile, with serious endocrinological and immune function side effects, excludes
them from life-long treatment of CP (Ballantyne and Mao 2003). The only exception
would be those rare patients who draw major benefit from not too high PO doses
of morphine (e.g., MS Contin) (Yamamoto et al. 1997: 2/39 BCP patients; Attal
et al. 2002: 3/15 BCP/CCP patients) or other congeners (e.g., methadone and
dextropropoxyphene, also NMDA antagonists) who are fully informed of long-term
consequences. As we have seen, the first patient in history to be diagnosed with

Drug Therapy

CP was opioid unresponsive (Edinger 1891; see Chapter 1) and, among others,
Davis and Martin (1947) found opioids ineffective for true CP of cord origin.
From a pathophysiological standpoint, opioid unresponsiveness may simply
depend on low
rather than loss of
opioid receptor binding in human SI (Pfeiffer
et al. 1982; Sadzot et al. 1990; Jones et al. 1991; Willoch et al. 2004): medial pain
system brain areas (i.e., thalamus, ACC, PFC, insula, temporal cortex and others)
have a high density of opioid receptors, and this would point to a critical role of
the sensory cortex in CP mechanisms (Chapter 8). However, morphine may have
some effect on non-thermal allodynia, a likely sensitization-driven event (Chapter 8).
Interestingly, opioids inhibit GABA interneurons, and, in light of the high efficacy
of GABA agonists, this would be further reason to limit their use. Opioid unresponsiveness of CP speaks against a functional impairment of the CNS opiate system.
Naloxone has not proved analgesic in a RCT (see Tables 5.1
5.4), although the dose
employed was lower than per initial claims of efficacy. One notes the impossibility
for an agonist and its antagonist to have therapeutic effects on the same disease, bar
invoking unknown mechanisms of action.
Cannabinoids have a mixed (short) track record up to now, but the impression
is that they will not represent a substantial advance over currently available drugs
(see also Warms et al. 2002).
f. Miscellaneous agents. The following agents had no impact in the treatment
of CP: NSAIDs, misoprostol (except for MS-associated trigeminal neuralgia),
calcitonin, IV adenosine (which triggers frequent headache) (Eisenach et al. 2003),
cholinergics (despite M1 and M4 and nicotinic receptors involvement in endogenous analgesia; Mullan 2002), atophanyl, phenotiazines (which have severe side
effects, including tardive dyskinesia), antihistaminics, mephnesin, IV tetraethylammonium (Davis and Martin 1947). Topical pain (band-like area or patch) may
be treated with an amitriptyline-ketamine or gabapentin cream, but there is no
published experience for CP.
We do not expect to achieve relief of CP with anti-absence drugs, such as
ethosuximide (a T Ca2 blocker), while levetiracetam gave no benefit in one personal
patient).
g. Combinations. These have never been addressed in a controlled manner.
However, in our experience, a combination of high-dose mexiletine and gabapentin
properly administered proved synergic and highly effective in more than 50% of
the patients we treated, with manageable side effects; as these were almost all
referral cases who did not respond to prior therapy (amitriptyline, carbamazepine),
this can be considered a successful combination (Canavero and Bonicalzi 2005).
In epilepsy, promising combinations have been found to include a sodium blocker
and a GABA agonist, two GABA agonists or two glutamatergic drugs with different
receptor profile (Deckers et al. 2000). Intriguingly, even for CP, a combination
of a sodium blocker (mexiletine) and a GABA agonist (gabapentin) proved particularly worthy. Antidepressants and antiepileptics in combination have not yet been
assessed in a controlled trial for CP. Contrary to animal data, we see no place for
combinations of morphine with gabapentin or dextrometorphan.

181

182

Central Pain Syndrome

2. Drug aggravation of CP
Some of the drugs reviewed worsened CP in a few patients: ketamine in 2 patients
reported by Yamamoto et al. (1997), propofol in 3 and dextrometorphan (100 mg)
in 2 of our patients (unpublished observations), but also IT baclofen (Loubser et al.
1996). These should not be dismissed, as they may point to focal neurochemical
mechanisms in single patients.
We would also caution against the use of modafinil, a glutamatergic agonist and
GABA antagonist used in the treatment of MS: this might worsen MS-associated CP.

NEUROMODULATION

Perhaps we can now envision a day in which, with the use of stimulation techniques, we
can take advantage of the brains natural modes of organization and reinforce them in time
of need, whether to control pain , . . . epileptic . . . discharge, or . . . tremor.
(Ervin and colleagues, 1966)

When oral or parenteral drugs fail, the problem is what to do. In view of the
continuing efforts aimed at neural reconstruction in the human brain and thus
physiologically revert pain, and progress in neuromodulation and drug therapy,
today there is only little room left for ablative procedures. Despite temporary initial
benefit with several of these, destructive surgery at any level of the CNS has only
a low long-term (45 years) success rate, with a high incidence of recurrence and
only few lucky patients totally relieved in the long term (Tasker 2001). Moreover, all
techniques carry a serious risk of permanent, disabling complications, including new
or worsening of pre-existent CP, as ablation only adds further damage.
The only true option is neuromodulation. This can be mainly achieved through
electrical stimulation of the damaged nervous system or intrathecal pharmacologic
infusion through implanted pumps.

ELECTRICAL
The analgesic effect of electrical stimulation has been known since Roman (and
perhaps even earlier) times, when the shock from an electric fish was used to relieve
gout pain and other pains, a custom that was not lost in the following centuries.
The eighteenth century witnessed a resurgence of this technique, despite strong
opposition. With Galvani and particularly Alessandro Volta in the nineteenth
century, electrotherapy was poised to make progress. One of the true pioneers in the
field was the Frenchman Duchenne de Boulogne, who published a classic book in
1855, in which he reported his experience in treating a wide variety of conditions,
and predominantly pain, with custom-made machines; Hermel too treated neuralgias
with electro-puncture (historical review in Sedan and Lazorthes 1978). Riddoch
(1938) noted that CP could sometimes be diminished by concomitant stimulation
183

184

Central Pain Syndrome

through faradization in the abnormal or adjacent normal body parts. Today, more
sophisticated means are employed.
1. Extra- and subdural motor (MCS) and sensory cortex
stimulation (PCS) (Table 6.1)
CP is the original disorder that led Tsubokawa and colleagues to try MI (BA4)
stimulation in 1988. Currently, the procedure is performed under local or general
anesthesia. A stimulating paddle is positioned on the dura overlying the motor or
sensory cortex trough two burr holes or alternatively a small craniotomy or craniectomy under fMR and/or neuronavigational conditions and, if effective parameters
are found after a test period, hooked up to a subclavear pacemaker. A subdural
approach may be elected in cases of pain involving the foot and distal leg. Permanent disabling morbidity (including epilepsy and intracerebral hemorrhage) and
mortality have not been reported. Stimulation parameters have been the most
diverse, analgesia having been obtained with both high and low frequencies (even
4100 Hz), low and high voltage, continuous and intermittent stimulation. The
choice of stimulation parameters is also dependent on the presence of the so-called
post-effect. Many, but not all, patients have their pain relieved or improved almost
immediately during intraoperative stimulation for periods ranging from several
minutes to hours or several days without further stimulation. This effect has a
tendency to abate over time and by the second month may stabilize at several
minutes to a few hours. Analgesia also can fade over time. Repositioning of the
electrode or intensive reprogramming may restore benefit in some cases, although
often at a lower level. Tolerance and fatigue are proposed mechanisms of such effects.
Best results are seen when the stimulating poles overlie parts of cortex corresponding
to painful body parts, although some data suggest that precise, millimetric, somatotopic localization of the electrode may not be required (see full-breadth review and
operative indications in Canavero and Bonicalzi 2002, 2003, 2004, 2006).
Analgesia can be expected when patients display mild (but not severe) sensory loss
particularly normal or only mildly elevated thermal sensory thresholds within the
painful zone and intact or almost intact corticospinal motor function, but several
exceptions exist making these relative criteria only (Canavero and Bonicalzi 2002,
2006). Most importantly, pharmacological tests can predict analgesia in the single
patient. Barbiturate- and/or ketamine-responsive patients can expect pain relief.
We found a good correlation between analgesia with CS and response to transcranial
magnetic stimulation (TMS) and between these two and propofol test (Canavero
et al. 2003); apparently, responsiveness to GABA marks patients that are particularly
favorable for stimulation. TMS also appears to predict response to CS, but, just like
CS, its analgesic effect is not as strong (Canavero et al. 2003, 2005), and even when
conducted daily for five consecutive days at higher frequency than usual (20 Hz), the
duration of effect tends to be short (less than 2 weeks), with patients unpredictably
obtaining excellent or no benefit (Khedr et al. 2005: 12 CPSP patients).
Even transcranial direct current stimulation does not appear to improve
substantially on these results, being particularly effective for paroxysmal pain
rather than the more resistant continuous pain (SCI CP: Fregni et al. 2006, 11 active
therapy patients).

185

CP (25 pts)

Thalamic pain (7 pts)

CPSP (12 pts; thalamic lesion: 6 pts;

putaminal lesion: 3 pts; pontine
hemorrhage: 1 pt; other lesions: 2 pts)

CPSP (11 pts; thalamic stroke: 8 pts;

putaminal hemorrhage small lesion
in the posterior limb of the internal
capsule): 3 pts)

Tsubokawa et al. (1991)

Tsubokawa et al. Acta Neurochir Suppl

(Wien) 1991; 52:1379

Tsubokawa et al. (1993)

(Also see Tsubokawa et al. Pain 1993;
58 (Suppl.):150)

Type of pain (no. of patients)

Tsubokawa et al. Pain 1990;

Suppl 5:S491 (abs.952)

Tsubokawas group

Author(s)

Intermittent stimulation effective in 5/12 pts. No

seizures; pain relief at stimulus intensities below
movements threshold. Paresis improvement. Pain
improvement in barbiturate-sensitive, morphineresistant pts. Disappearance of the analgesic
effect in 3 pts, with reappearance after revision of
electrode placement
Barbiturate-sensitive pts: 5 (3?)/11;
morphine-resistant pts: 10 (1?)/11.
Stimulation of area 4 ipsilateral to the inciting
lesion. One week test period. Fair and poor
responders not implanted. Satisfactory immediate
pain relief in 8/11 pts (73%). Gradual effect
reduction over several mos in 3/8 pts. Long-term
response in barbiturate-sensitive and morphineresistant pts (also for Vc DBS). No pain-relieving
effect by high-frequency postcentral stimulation
(11/11 pts): in 2, worsening of pain, similar to
their spontaneous ones. In 3 pts, areas rostral to
MI stimulated without relief. Nonpainful
paresthesias unrelieved

Complete pain relief in 5/12 pts (1 yr),

considerable pain reduction in 3/12 pts
(1 yr). Long-term benefit in 8/12 pts
(41 yr)

Pain relief:
Immediate:
Excellent (480%): 6/11 (54%)
Good (6079%): 2/11 (18%)
Fair (4059%): 1/11 (9%)
Poor (<40%): 2/11 (18%)
Long-lasting:
Excellent (480%): 5/11 (45%)
Good (6079%): 0/8
Fair (4059%): 0/8
Poor (<40%): 3/8 (37%) (2 yrs)

(continued)

MCS improved movements of the painful limbs.

Pain subsided within a few minutes. 510 min
ON, 45 h relief 56 times daily, then
23 times daily

First report of MCS for CP (1988) MCS vs.

thalamic stimulation. MCS is more effective
than thalamic stimulation. Improvement of
motor function in some pts

Notes

Excellent (5: no drugs needed) or good

(2: some drugs needed) pain relief

Effective in 75% of cases (47 mos)

Results (follow-up)

TABLE 6.1. Motor cortex stimulation (MCS) (arranged by reporting groups)

186
Pain relief 440% in 1 pt previously
unsuccessfully treated by Vc DBS. No satisfactory
pain control by thalamic stimulation in any pts
Suprathalamic stroke infarct or hemorrhage of
the posterior limb of internal capsule, or parietal
lobe, sparing the thalamus. No pts with midbrain
or medullary lesions. MCS test period: 1 wk.
8/39: morphine responsive (M)
22/39: thiamylal responsive (T)
11/23: ketamine responsive (K)
Thiamylal ketamine sensitivity morphine
resistance may predict a positive effect of MCS

Damage of the posterior limb of the internal

capsule in pts with putaminal hemorrhage.
Previous ineffective SCS. Pain relief 460%:
13/18 pts (73%) with no or mild motor weakness
(70% of pts with inducible muscle contraction);
2/13 pts (15%) with moderate or severe motor
weakness (difference statistically significant).
Satisfactory pain control in 14/20 pts (70%) with
inducible muscle contraction, but in only 1/11 pts
(9%) without inducible motor contractions
(p < 0.01). No relationship between pain control
and presence of hypesthesia, dysesthesia,
hyperpathia, allodynia or disappearance of SSEP
N20 wave plus stimulation-induced paresthesias,
or motor performance improvement. 3 pts with
MCS or DBS became pain-free without stimulation
for years (all 3 getting initial excellent relief at

MCS in 3 pts. Pain relief: 2/3 4 60%;

1/3 4 40% (4 mos)
28 MCS.
Excellent/good (50100%)
pain relief:
Thalamic pts: 10/19 (53%)
Suprathal. pts: 3/9 (33%) (difference
not significant)
T or K & M: 2/4 (50%)
T or K & M
: 10/14 (71%)
T
& K
& M: 0/2 (0%)
T
& K
& M
: 1/8 (13%)
Overall: 13/28 (46%) (12 mos)
Early satisfactory (460%) pain relief:
23/31 pts (74%).
Long-term efficacy
(2 yrs): 15/31 pts (48%)

CPSP (6 pts; lateral medullary infarct)

CPSP (39 pts; thalamic stroke:

25 cases; suprathalamic stroke: 14 pts)

CPSP (31 pts. Thalamic stroke: 20 pts;

putaminal hemorrhage: 8 pts; lateral
medullary infarction: 3 pts)

Katayama et al. Stereotact Funct

Neurosurg 1994; 62:2959
(Tsubokawas group)

Yamamoto et al. (1997)

(Tsubokawas group)

Katayama et al. (1998)

(Also includes: Katayama et al.
Stereotact Funct Neurosurg 1997;
69:739)
Tsubokawa et al. In: Abst. 3rd
Int Congress INS.
Orlando, 1996,
p. 123)

Some of the pts already reported in

Tsubokawa, J Neurosurg 1993, above
(Tsubokawas group)

Notes

Results (follow-up)

Type of pain (no. of patients)

Author(s)

TABLE 6.1 (continued)

187

Meyerson et al.
Acta Neurochir Suppl (Wien) 1993;
58:1503

Karolinska group:

Hosobuchi (1993)
Also includes:
Stereotact Funct Neurosurg 1992;
59:7683
Abstr. IASP Congress 1993

Parrent and Tasker (1992)

Tasker et al. 1994
Tasker 2001

Toronto group:

CPSP (3 pts; thalamic hemorrhage:

2 pts; brainstem infarction: 1 pt)

CPSP (2 pts)
CPSP (5 pts; post-removal of parietal
cortical AVM: 1 pt; brainstem infarction:
1 pt; thalamic lesion: 3 pts)

CPSP (1 pt, large suprathalamic infarct)

CPSP (31 pts)

Fukaya et al. (2003)

Also includes:
Katayama et al.
Acta Neurochir Suppl. 2003;
87:1213
(Tsubokawas group)

Katayama et al. (2001)

(Tsubokawas group)

Small thalamic stroke, then action

tremor, then Vim DBS, then
cardioversion, then CPSP
CPSP (45 pts)

Yamamoto et al. (2000)

Pain relief: CPSP: none: 3/3

Substantial pain relief with ipsilateral to

pain MCS with gradual abatement over
6 yrs; relief with subdural stimulation
over a few months.
1 relief, 1 failure
Pain relief:
Initial: 5/5 excellent.
At 23 mos: 4/5 excellent (450%);
fair. At 930 mos: 3/5 excellent
(thalamic, parietal, brainstem)

(continued)

In spite of multipolar electrode grid in 1 and

relocation of paddle in another

Efficacy dramatically reduced in 2 thalamic

pain pts, to 0% in 1 pt and 30% in 1 pt 26 mos
after implantation

Contralateral MCS due to a lack of sufficient MI on

the affected side. Stimulation-induced ipsilateral
paresthesias

Experimental study on conscious somatosensory

response during surgery for electrode placement

(See text) Sometimes SCS and DBS produced

long-lasting pain-free intervals (stimulator
switched off). DBS and MCS in 4 pts: better result:
MCS 1/4 pts; DBS 2/4

Satisfactory pain control:

SCS: 7% of pts;
DBS: 25% of pts;
MCS: 48% of pts
Unsuccessful MCS in 2 CPSP pts
reporting abnormal pain sensation after
stimulation of the motor cortex
(see text)

Thiamylal/ketamine responsive, morphine resistant

50% relief over 2 years

progressively longer stimulation intervals during

intermittent stimulation)

188

Nguyen et al. Arch Med Res 2000;

31:2635
Nguyen et al.
Neurochirurgie 2000; 46:48391

CP (16 pts; CPSP: 11 pts; thal.

abscess: 1 pt; head trauma: 1 pt; SCI:
3 pts)

Pain relief:
BCP:
Good (70100%): 5/13 brain; 1/3
SCI

Progressive loss of effect in some pts

reversed in most by correct repositioning of
paddle.

Pain relief in 13 pts described as CP by authors:

5 good (70100%), 5 satisfactory (4069%),
3 poor (<40%). SCI CP: 1: 70100% relief, 2:
<40% relief. Progressive loss of effect in 6/32 pts
(19%), due to electrode malpositioning in 5 pts.
After electrode repositioning, good pain relief. RCT
in 5 pts; 5/5 positive response
Satisfactory/good (40100%) pain
relief:
parietal stroke: 2/2 pts; deep brain
hematoma: 4/5 pts; thalamic stroke:
4/4 pts; thalamic abscess: 1/1 pts;
brainstem lesion: 1/1 pts; head trauma:
2/2 pts; SCI: 2/3 pts (350 mos)
(Mean follow-up 27.3 mos)

and both following papers:

Complete tremor relief

Complete pain relief (32 mos)

Facial pain arm tremor after removal

of an acoustic neurinoma (1 pt)
CP (parietal stroke: 2 pts; deep brain
hematoma: 5 pts; thalamic stroke:
4 pts; thalamic abscess: 1 pt;
brainstem lesion: 1 pt; head trauma:
2 pts; SCI: 3 pts)

Nguyen et al. Mov Disord 1998;

13:848
Nguyen et al. (1999)
(Includes all pts from previous paper)
Virtually the same patients reported in
Nguyen et al. In: Proc 4th Int Congress
of INS, Luzern, 1998,

Disappearance of allodynia in 2, but reduction only

of spontaneous pain in 1 pt in the thalamic pain
group. Globally, continuous pain improved in
4/10, evoked pain in 6/8 and paroxysms in 6/7.
Reduction of the initial pain relief (to less than
40% in 2) in 3 cases. Previous ineffective Vc DBS
(basal ganglia, deep lesion). Best results in the
parietal lobe infarct, the thalamic abscess, and in
one thalamic infarct pt (almost normal life, drugs
markedly reduced).
In sum: 58% of pts getting 440% relief

Notes

Pain relief:
(a) parietal lobe infarct
80100%: 1/1 pt (14 mos)
(b) basal ganglia hematoma
<40%: 1/3 pts (30 mos)
4060%: 2/3 pts (25 32 mos)
(c) thalamic pain
<40%: 2/5 pts (18 22 mos)
4060%: 1/5 pt (28 mos); 6080%:
2/5 pts (22 30 mos)
(d) post-traumatic lesion
<40%: 1/1 pt (29 mos)
(e) SCI
<40% in a paraplegic pt (18 mos);
100% (visceral pain substantial
reduction of diffuse pain) in a
tetraplegic pt (22 mos)

Results (follow-up)

CP (12 pts; parietal lobe infarct: 1 pt;

basal ganglia hematoma: 3 pts;
thalamic pain: 5 pts (4 stroke,
1 abscess); deep post-traumatic brain
lesion: 1 pt; SCI: 2 pts)

Type of pain (no. of patients)

Nguyen et al. Acta Neurochir Suppl

(Wien) 1997; 68:5460

Paris group:

Author(s)

TABLE 6.1 (continued)

189

[includes: Nuti et al. (2005)

Peyron et al. Pain 1995; 62:27586
Garcia-Larrea et al. Stereotact Funct
Neurosurg 1997; 68:1418
Garcia-Larrea et al. (1999)
Mertens et al. Stereotact Funct
Neurosurg 1999; 73:122
Sindou et al. 9th World Congress on
Pain, Book of abstracts, IASP Press,
1999
Montes et al. Neurophysiol Clin 2002;
32:31325

Lyon group:

Drouot et al. (2002)

Same pts as reported above

19922003

CP: 27 pts.
Ischemic lesions: 11 pts (3 thalamic
(Vc), 4 medulla, 2 cortical parietal, 1
parietal/insula/ACC, 1 parietal/insula).
Hemorragic lesions: 11 pts (1
thalamic (Vc), 1 thalamus/midbrain,
5 capsulothalamic, 1 capsulolenticular/
insula, 3 cortical parietal).
Frontoparietal trauma: 1 pt.
SCI (discal hernia-associated
myelopathy): 3 pts. Spinal conus AVM:
1 pt.

CP (13 pts; parietal stroke: 2 pts; thal.

stroke: 8 pts; thal. abscess: 1 pt; SCI:
2 pts)
19932000

(continued)

Prospective evaluation of MCS. Long-term outcome

evaluated by means of: 1) % pain relief, 2) VAS,
3) postoperative VAS decrease, 4) reduction in
drugs intake, 5) yes/no response to being
operated on again.
MCS efficacy not predictable by motor status, pain
characteristics, lesion type, QST, SSEP/LEPs, pain
duration, BCP vs. CCP, presence of evoked pain.
No subjective sensations during active stimulation.
Partial epileptic seizures in 3 pts in the early
postoperative stage or during trials for increasing
intensity. 1 speech disorder and 1 motor deficit
resolved spontaneously. Long-term relief predictable from early pain relief.
12 paddles.
Parameters: 0.55 V, 3080 Hz, 60330 ms
MCS may have adverse cognitive effects. The risk
may increase with age (450 yrs)

MCS effective in pts with normal or quite normal

non-nociceptive thermal threshold within the
painful area or in pts with improved sensory
thresholds during MCS. QST testing bilaterally; in
non-responders, MCS induced significant changes
on the side of stimulation.
Parameters: 3 h ON, 3 h OFF, 60 ms, 40 Hz,
13 mV

Good pain relief (440%): 9 (69%)

Unsatisfactory pain relief (<40%): 4

Follow-up: 2104 mos

(mean 49 mos).
Pain relief:
BCP:
excellent (470%): 3; good (4069%):
8; poor (1039%): 8; negligible
(09%): 4
CCP: excellent: 0; good:
3; poor: 1; negligible: 0.
Decreased analgesic intake: 52% of pts
(complete withdrawal 36%).
Unchanged: 45% of
pts. Unavailable data: 3%. Decrease/
withdrawal of analgesic in 10/11 poor
responders (contradictory results as
noted by the authors).
Favour re-intervention: 70% of pts

In sum: 66% had 30100% relief,

but less with at least 50% relief

Satisfactory (4069%): 5/13 brain

Poor (<40%): 3/13
Substantial pain relief:
BCP: 10/13 pts (77%)
SCI: 1/3 pts (33%)
(350 mos; mean
follow-up 27.3 mos)

190

Type of pain (no. of patients)

Saitoh et al. Neurosurg Focus 11(3),

article 1, 2001
Includes Saitoh et al. J Neurosurg
2000; 92:1505

Pain relief:
Immediate long-term (mos)
CPSP (8 pts):
Excellent 1 [1] (3); Good: 1[1]
(48) (putam.);
Fair: 2 [2] (4, 25);
Poor: 4 [not implanted with IPG]
SCI:
Excellent: 1 1 (1)

CP (CPSP 5 pts; SC pain: 2 pts) 1

(algodystonia)
19932003

Canavero and Bonicalzi (2002, 2003)

Includes: Canavero et al.: In: Proc 4th
Int Congress of INS, Luzern, 1998
Canavero et al. Neurol Res 2003;
25:11822

CPSP (thal. stroke: 6 pts; putaminal

hem: 1 pt; pontine hem: 1 pt) SCI (1 pt)
19962001

Effective (30100%) pain relief with

MCS/PCS in 2/7 pts. Long-term
efficacy (4 yrs) in 1 pt (MS CP).
Ineffective MCS in 4/7

CPSP (1 pt; thalamocapsular stroke)

Canavero et al. (1999)

Saitohs group:

Effective short-term pain relief

(allodynia disappearance and 50%
reduction of burning pain) (5 wks)

CP (2 pts; CPSP: 1 pt; syringomyelia:

1 pt)

Pain relief: 3050% in syringomyelia pt

(2 yrs); no relief in CPSP

Results (follow-up)

Canavero and Bonicalzi (1995)

Turin Advanced Neuromodulation Group (TANG)

Author(s)

TABLE 6.1 (continued)

Pain relief categorization:

Excellent (80100% relief)
Good (6079% relief)
Fair (4059% relief)
Poor (<40% relief)
Poor responders not implanted. Pharmacological
test with phentolamine, lidocaine, ketamine,
thiopental, morphine, placebo. Reduction over the
time of MCS pain relieving effect in 3/10
implanted pts.
Some pain reduction by SI
stimulation. Ineffective prefrontal stimulation.
Subdural placement in some pts (5),

Effective SI stimulation in 1, then resubmitted to

MCS with same benefit plus 50% opioid reduction
(however, patient unsatisfied and explanted).
Overall efficacy: 3/7 CP pts, all propofolresponsive. Ineffective MCS in 4/7 CP pts, all
propofol-unresponsive, but 1 who could not be
assessed due to intermittent nature of pain.
Algodystonia: temporary benefit

Propofol-responsive pt. Painful supernumerary

phantom arm during MCS and lasting 6 mos after
stimulator switch-off. Pain relapse after 5 wks

Syringomyelia pt: parietal somatosensory

stimulation. Spreading of pain to contralateral side
and vanishing of analgesia at 2 mos. Modest
propofol response; CPSP pt: propofol
unresponsive

Notes

191

Cioni et al. Proc XLV SINCH Congress,

1996

CP (4 pts; thalamic pain)

Pain relief (5060%): 1/4 pts, but

unsatisfactory relief at 1 yr

(continued)

Extradural multipolar (1620) grid in all plus

electrophysiologic mapping; several combinations
assessed over 12 h

The only pt where it was tried: propofol-sensitive.

Pain disappearance for 5 mos after stimulator
switched off in the responder. Enduring benefit in
1 pt only

Appreciable pain relief:

1 pt, cortical (4 yrs);
2 pts (weeks to months)
No pain relief: 4/7 pts (thalamic,
brainstem)

Brainstem injury: 5060% (31 mos): 1
pt

CPSP (cortic. stroke 1 pt; thal. stroke:

3 pts; brainstem stroke: 2 pts). Gunshot
brainstem injury (1 pt)

Nandi et al. (2002)

Includes all pts reported in Carroll et al.
Pain 2000; 84:4317
Smith et al. Neurosurg Focus 2001;
11(3):article 2

Other groups:

PET study (see text)

Immediate successful pain relief (VAS

decrease from 8 to 1)

CPSP (1 pt)

Modified MCS protocol: subdural MCS within the

central sulcus. Implants in interhemispheric
fissure: 5 pts (lower limb pain); within central
sulcus: 5 pts (area 4 and area 3b stimul.)
surface of the precentral gyrus.
Area 4 within central sulcus seems to be the
optimal stimulation point.
Ketamine-sensitive pts seem to be good
candidates for MCS

Saitoh et al. (2004)

Likely included in series above
Oxford group:

Overall pain relief:

Excellent: 6/19 pts
Good: 3/19 pts
Fair: 5/19
None: 5/19

CP (11 pts; CPSP: 9 pts; brainstem

injury: 1 pt; SCI: 1 pt)

Saitoh et al. (2003)

Includes all
previous pts

Tani et al.
J Neurosurg 2004; 101:6879

interhemispheric in 3. Dual IPGs driving 2 paddles

in 2 pts
First report of bilateral cortical stimulation
for SCI pain. 4 mos interval between implants.
Thiamylal, lidocaine, ketamine, phentolamine,
morphine, and placebo unresponsive pt. Pain
control with 34 periods (30 min each) of
stimulation a day, followed by 56 h benefit
without stimulation. Left pain relapse after right
system removal (infection, 6 mos after surgery)

192

Thalamic pain (2 pts)

CPSP (2 pts; putaminal hemorrhage: pt

A; post-20 mos stereotactic
thalamotomy: pt B)

CPSP (thal. infarction: 2 pts; thal.

hemorrhage: 5 pts)

CPSP (1 pt: capsuloinsular hemorrhage)

CPSP (evacuated putaminal hematoma)

(1 pt)
SCI (1 pt)

SCI CP (1 pt)

Herregodts et al. (1995)

Migita et al. (1995)

Fuji et al. (1997)

Barraquer-Bordas et al. (1999)

Kuroda et al. (2000)

Mogilner and Rezai (2001)

Rodriguez and Contreras (2002)

MCS ineffective. Later SI/SII CS

effective for 4 yrs
Relief (not broken down) (mean followup 6 mos)
Evoked pain dramatically improved.

MCS trial ineffective (motor response

elicited)

Satisfactory pain relief: 6/7 pts (1 mo)

Unsatisfactory pain relief: 5/7 pts
(3 mos)

Immediate pain relief: 450% in both

pts. Long-lasting: 1/2 (full relapse in
one at 4 mos)
Pain relief: 7080% in pt A (1 yr)
No relief in pt B

70% pain relief in 1 thal. pt. (3 yrs)

Gradual abatement of pain relief over
2 yrs. 6090% relief, then 5070%,
then 2030% at 341 mos (ave: 27
mos)
Satisfactory (3050%) pain relief:
Pts. A (44 yrs) and B (42 yrs).
Short-term pain relief (6 mos): pt C

CPSP (thal. stroke: 2 pts; brainstem

stroke: 1 pt)

Dario et al. Long-term results of chronic

MCS for CP.A185 In Abstr. 9th World
Congress on Pain, IASP Press, 1999
Also includes Dario et al. Riv Neurobiol
1997; 43:6259
Franzini et al. (2003) Also includes
Franzini et al. In: Abstr. XLVIII Congresso
SINCH, Copanello, 1999
Franzini et al. J Neurosurg 2000;
93:8735

CPSP (3 pts, A, B, C)

Results (follow-up)

Type of pain (no. of patients)

Author(s)

TABLE 6.1 (continued)

SCI pain following cervical ependymoma removal.

Third party analysis of results. Tremor improvement.

Hemisoma burning pain, evoked pains. DBS

reduced CP for 5 mos and evoked pains, until
glioma displaced electrode with relapse and death

Lesions included internal caspule, Vc and pulvinar

(MRI confirmed, 5 pts). Early electrode removal in
1 pt after unsatisfactory test stimulation

Pt A: morphine and barbiturate unresponsive. 30%

pain relief with TMS
Pt B: previous 6 mos effective Vc DBS.
Barbiturate responsive, morphine and TMS
unresponsive

2 responders propofol-sensitive. Pain abolition

after a second stroke in pt B.
Unsatisfactory pain relief (30%) by further stimulation in pt C.
Complete abolition of thalamic hand

All pts propofol-responsive. 22.5 V, 5075 Hz,

120210 ms, continuous mode.

Notes

193

CPSP (1 pt)

CPSP (1 pt)

CPSP, Wallenberg (1 pt)

CPSP, thalamic (1 pt)

CP (1 pt)

CPSP, brainstem (1 pt)

Frighetto et al. (2004)

Henderson et al. (2004)

Brown and Pilitsis (2005)

Savas and Kanpolat (2005)

Slawek et al. (2005)

20% reduction on VAS; withdrawal of

narcotic and decrease of non-narcotic
medications, ability to introduce
rehabilitation and improvement of sleep

No relief

Relief, then loss, then new relief (?)

after intensive reprogramming
0%;
VAS 10 to 8; McGill Quest. Index
from 65 to 32 (both sensory and
affective scores)

Relief (no details given)

Steady burning pain moderately

relieved (2 mos)

Follow-up: 4 mos. No side effects

(continued)

Follow-up max. in whole series (PNP and CP):

10 mos.
Contrary to Nguyen, they conclude that
precise, somatotopic localization of the electrode
may not be required, because the optimal
interelectrode distance determined during cortical
mapping and afterwards with subjective patient
evaluation of pain control was fully 3 cm.
Intraoperative neuronavigation and cortical
mapping for stimulation site targeting. Strength
and discriminative sensation improvement from
MCS in 3 pts with facial weakness and sensory
loss. Dysarthria improvement in 1 pt
More than 50% reduction in pain medication dose
They state that some experienced authors have
expressed to them dissatisfaction with MCS. MCS
performed at many more centers than those which
publish; most of the failures go unreported and
only series with good results are published

Previous ineffective thalamotomy

No reduction of analgesic intake after MCS. 7.1 V,

5 Hz, 450 ms, ON 2 h, OFF 3 h, 0
/2

194
3 responders (31%, 41%, 62%)
2/7 pts placebo responder.
Duration of positive effect: 2, 4,
1.5 years (4.56.0 V, 5085 Hz,
210250 ms).

CPSP (thalamic): 7 pts

19942005

Rasche et al. (2006)

Includes: Tronnier VM. Schmerz 2005;

15: 2789

Relief of dysesthesia, allodynia and

hyperpathia in 2 CPSP pts (pts were
able
to touch the painful area without having
painful sensations).

5075% drug dosage reduction among

responders.
Third party evaluation.
Plegia not an unfavorable prognostic factor.
Study evaluating the usefulness of the combination
of fMRI and intraoperative cortical brain mapping
(iCM) as functional targeting methods for MCS

Pain relief (%)

100%/50%/worsening
83%/ failure (both plegic)
87.5%
100%
70% Failure

CPSP, subcortical (3 pts);

capsular (2 pts);
brainstem (1 pt);
MS pain (1 pt);
cervical syrinx (1 pt);
SC ependymoma (1 pt)
19982003

Pirotte et al. (2005)

Also includes Pirotte et al. Neurosurg
Focus 2001; 11(3)

Test trial including a double blind test. Results

evaluated by means of VAS. Mean follow-up 3.6
years (range 110 years).
Single burr hole, neuronavigation.
No sensation evoked by stimulation. All responders
on continuous stimulation. After implantation,
intermittent stimulation.
Lasting pain reduction with minor changes of the
stimulation parameters. Immediate or almost
immediate (3060 min) pain reduction after
turning the MCS on. MCS effect lasting from
30 min up to several hours after cessation of the
stimulation

Frameless neuronavigation. Single burr hole

and vacuum headrest. Awake patient. No
complications.
Third party evaluation.
Description of an integrated protocol for precise
electrode placement (functional image guidance:
volumetric rendering of a 3-D MR data set with
superimposed fMR imaging data plus
intraoperative electrical stimulation)

70100% relief
(follow-up: 618 mos)
90% relief (follow-up: 24 mos)

CPSP (hemorrhage) (3 pts)

CP, insular (1 pt)

Gharabaghi et al. (2005)

Also includes Tirakotai et al.
Minim Invasive Neurosurg 2004;
47:2737

Notes

Results (follow-up)

Type of pain (no. of patients)

Author(s)

TABLE 6.1 (continued)

Neuromodulation

CS is the only electrical neuromodulatory technique which allows for blinded,

placebo-controlled assessments: CS does not generally induce any motor activation,
even at high voltage, or any sensory phenomena in a majority of patients. A pure
placebo effect has been excluded (Canavero and Bonicalzi 2002; Rasche et al. 2006).
In a few cases, though, sensory effects can be evoked with both MCS and PCS,
but are unrelated to eventual analgesia. Although no analgesia was seen in the
initial series of Tsubokawa of CS with SI stimulation, PCS has been confirmed to be
analgesic in CP (see Table 6.1).
More than half of BCP patients gain 40% or more relief at 4 years. A suggestion of
greater response of evoked versus spontaneous pains (see Table 6.1) is not confirmed
in most series and MCS does not relieve nonpainful paresthesias.
Worsening of the original pain via ipsi- or contralateral stimulation of MI/SI
has been reported in a few patients, and one of our CP patients developed a painful
supernumerary phantom arm after MCS (see Canavero and Bonicalzi 2002).
Tsubokawas group also reported a few patients with excellent analgesia and
increasing periods of post-effect who, after having the stimulating apparatus switched
off in 2 years, never had their pain return, interpreted as a sign of neuroplastic
phenomena induced by MCS in SI (1 case also in Peyron et al. 1995). Analgesia via
ipsilateral stimulation is on record. Nguyen and colleagues (see Table 6.1) saw no
major modification of somatotopic arrangement in their patients, but one obtained
bilateral benefit from unilateral stimulation. Rainov and colleagues (see Table 6.1)
found that, by changing the polarity of the electrodes, it was possible to induce
tingling sensations and muscle activation not only contralaterally to the stimulated
MI but also in the ipsilateral part of the face.
Mechanism of action. Tsubokawas original hypothesis (based on animal experiments) that MI, but not SI, stimulation restores the inhibitory surround of
hyperactive SI pain-coded cells by anti-/orthodromic activation of non-nociceptive
neurons has been disproved by successful cases of PCS and he later rejected it.
Nonetheless, Drouot and colleagues (2002) concluded that MCS reinforces the
control of non-nociceptive sensory inputs on hyperactive nociceptive SI cells, at least
when these sensory afferents are partially preserved (implying that lemniscal fibers
inhibit STT fibers), with improvement of sensory discrimination. In poor responders,
MCS did not modify the sensory thresholds measured within the painful area, but
induced significant changes on the opposite side, i.e., ipsilateral to stimulation. Lack
of effect was ascribed to more severe disturbances of the sensory systems. However,
the literature reports patients with impaired lemniscal transmission who were
relieved by MCS (see Table 6.1) implying that a normal lemniscal system (gate
control) is not required to obtain good results. We have seen CP patients relieved
by propofol, who also had restoration of normal sensation (see Canavero and
Bonicalzi 2004). Parenthetically, MI is actually one of the areas activated after
acute noxious stimulation in humans (Coghill et al. 1999).
Some authors believe that MCS does not act at cortical levels below the electrode, but through descending axons. For these, pyramidal neurons or their efferents
(perhaps relaying to SI or thalamus) are important, as these can be activated even at
intensities below the muscle contraction threshold; previously, Fields and Adams

195

196

Central Pain Syndrome

(1974) stimulated pyramidal fibers in the internal capsule to relieve human pain,
as they believed that CP is due to loss of intracortical inhibition, with relative
preservation of excitation. However, in humans, there are few descending fibers
from SI to the dorsal horn, and these do not end in laminas IIII (Schoenen and
Grant 2004). Similarly, there are no or few descending fibers from MI to the
superficial dorsal horn in man (Schoenen and Grant 2004). It is difficult to believe
that CS in humans acts by descending direct inhibition to the spinal cord (see also
Meyerson and Linderoth 2001). Other authors believe a cortical effect is foremost
in this regard.
Neurometabolic studies of this problem have been published. Garcia-Larrea and
colleagues (1997) studied 7 CPSP and 3 PNP (brachial plexus avulsion, BPA, pain)
patients submitted to contralateral MCS (in 3 medially, i.e., subdurally). H2(15)O
PET was done before, during 5 and 20 minutes and 30 minutes after a 20 minute
session of stimulation. Results were not differentiated between CP and BPA. There
was no significant difference in regional cerebral blood flow (rCBF) between the
two controls or the two stimulation conditions. The only locus of significant CBF
increase during MCS was observed in the motor thalamus. Sizable, but insignificant,
CBF increases during MCS were seen in the left insula, BA2432 and upper
mesencephalon (plus a rCBF decrease in BA1819 bilaterally). No significant change
was seen in MI (SI could not be resolved with their machine). All changes were
reversible upon stopping MCS, although BA24 and mesencephalic changes persisted
or even increased slightly after stoppage of MCS. They compared 3 patients with 80
100% relief and 4 with less than 40% relief. Mean thalamic CBF was enhanced in
both groups with a similar time course, albeit rCBF increase was greater in those with
80%-plus relief. In contrast, mean CBF in BA24-32 appeared to increase during
MCS only in patients with good relief and to decrease in poor responders, even in
individual analyses. The same group (Laurent et al. 1999; Garcia-Larrea et al. 1999)
evaluated 10 patients with CP and BPA (likely including the above-mentioned
patients, although time from implantation to PET does not correspond). MCS was
stopped 24 hours before PET. Four consecutive scans were first recorded (A). Then
PET was recorded at 5, 15, 25 and 35 minutes after switching on MCS (B). MCS was
subsequently stopped and PET recorded at 15, 30, 45, 60 and 75 minutes after MCS
had been turned off (C). MCS (B versus A) was associated with increased rCBF in
rostral ACC contralateral to the electrode. During MCS stoppage (C versus A)
there was strong activation up to 75 minutes after MCS discontinuation of rostral
ACC, orbitofrontal cortex, basal ganglia and brainstem. MCS (BC versus A) was
associated with decreased blood flow (suggesting constriction) on the dura
immediately below the electrode. Images of CBF changes in the brainstem did not
cover the localization of the PAG. They did not find MCS activation of SI, a likely
consequence of the spatiotemporal resolution limits of their PET machine. The lowthreshold analysis (Z-score 3.5) of the two-step procedure yielded some regions of
significant CBF increase: the whole thalamus (ipsilateral to MCS), the ACC (mostly
contralaterally to MCS, plus midline), orbitofrontal areas, a region comprising the
insula and descending towards the inferomedial temporal lobe including amygdala
(exclusively contralateral to MCS), subthalamic-upper brainstem region (ipsilateral
to MCS). The second (high-threshold) step of the analysis (Z-score 4) restricted

Neuromodulation

spatially the above results and limited the anatomical region of significant CBF
increase to thalamic VL ipsilateral to MCS, with extensions to VA and subthalamic
region. Vc was outside the region of increased CBF in both high- and low-threshold
analyses. The sequence included condition A (CBF assessed basally, 15 minutes
before MCS with stimulator turned off for 18 hours), conditions B and C
(2 consecutive scans performed respectively after 5 and 20 minutes of continuous
MCS) and condition D (scan after 30 minutes after MCS discontinuation). Pain
ratings during PET were 4.8 + 2.6 during condition A, 4.3 + 2.9 and 3.69 + 2.8 in
conditions B and C and 3.69 + 2.8 in condition D. In spite of a trend to pain
decrease from A to D, differences were not significant. As far as rCBF changes are
concerned, in all cases there was an abrupt CBF increase during the first scan under
MCS (5 minutes after onset) which remained stable during PET 20 minutes after
MCS onset. These effects were reversible 30 minutes after MCS interruption in all
sites, except in ACC where rCBF had not yet reverted to pre-stimulation values
30 minutes after MCS discontinuation: here two spots of increased rCBF appeared in
right and left ACC/orbitofrontal boundaries and stayed almost so after switching off
the stimulator. No significant change related to MCS was observed in SI or MI. CBF
decreased in BA1819 areas and were totally reversible upon discontinuation of MCS.
In CP and BPA patients with 480% versus <20% relief, while lateral thalamic CBF
appeared to increase in all patients (albeit to a greater extent 15% versus 5% in
those relieved), BA32 CBF increased in responders (5% at 20 minutes), but
decreased in nonresponders (
10% at 20 minutes); upon close scrutiny, this does not
seem a strong finding as in their two reported CP cases this was not the case. These
studies suffer from limited statistical power due to small number of patients and
shortcomings of ROI measurements (analysis was based on ellipsoidal ROIs placed
over the lateral thalamus and BA32, with other regions not studied due to their
irregular shape) and the authors themselves considered their results exploratory and
deserving confirmation. Also, VAS values reported here correspond to those obtained
during the week preceding PET, and thus may have not fully reflected current pain
intensity. They (Garcia-Larrea et al. 1999) also recorded CO2 laser-evoked potentials
(LEPs) and flexion nociceptive reflex (RIII) in these same patients. LEPs (amplitude
and latency of each component) and RIII (surface) were studied with MCS turned
off, on and at least 30 minutes after MCS interruption. LEPs were obtained after
stimulation of both the painful and the intact side, while RIII was obtained after
stimulation of the painful side only. In one patient, after stimulation of the nonaffected side, LEP amplitudes of the vertex component decreased significantly during
active stimulation. In the group as a whole, after stimulation of the non-affected
side, LEP amplitudes tended to decrease under MCS, although not statistically
significantly. RIII was not modified in the three conditions. Electrophysiological
responses did not correlate with VAS. There was a lack of any significant acute
change in SEPs during MCS in any of the recorded patients with central lesions.
None of the 4 patients whose nociceptive reflexes remained unmodified by MCS was
satisfied with the attained analgesia. Although the 7 patients with CP had sizable
epidural SEPs during intraoperative monitoring, only 4 retained scalp-recorded SEPs
of enough amplitude to permit assessment of MCS effects. Parietal somatosensory
responses up to 50 ms post-stimulus did not exhibit any significant change in

197

198

Central Pain Syndrome

amplitude, latency or topography in relation to MCS. Thus, significant modulation

of spinal nociceptive reflexes was seen during MCS in 3/7 patients, while it was
unchanged in 4. Modification thereof corresponded in every case to attenuation of
the responses during MCS. Two of 3 patients with MCS-related reflex attenuation
experienced good to very good relief, while the third reported 460% abatement of
allodynia during MCS, but only 30% of spontaneous pain. These data too do not add
substantially to our understanding of how CS works.
Saitoh and colleagues (2004) submitted a right-sided CPSP patient to MCS, with
excellent analgesia (VAS 8 to VAS 1) after 30 minutes of stimulation. H2(15)O PET
pre- and post-stimulation revealed significant rCBF increases in left frontal areas
(BA9 and 11, BA32) and the left thalamus and decreases in temporo-occipital areas
(right BA22 and left BA19). The efficacy of MCS was mainly related to increased
synaptic activity in the thalamus, whereas all other changes were related to emotional
processes.
We (Canavero and Bonicalzi 1995; Canavero et al. 1999) and Tsubokawas group
(Tsubokawa et al. 1991) found that cortical stimulation changes both local cortical
(SI/MI) and thalamic rCBF with pre- and post-stimulation SPECT. F-MR studies
showed MCS to have inhibiting effects on SI/MI cortex as well as contralaterally,
supporting a cortical mechanism of analgesia (see references in Canavero and
Bonicalzi 2002). Moreover, subdural MI stimulation appears to activate axons in the
cortex, which excite both corticospinal neurons and local inhibitory neurons. The
effects are greater with cathodal stimulation (Hanajima et al. 2002).
We may conclude that CS may act locally by modulating (unscrambling) the
MI/SI dipole and the long thalamocortical reverberating loop, with subsequent fallout effects on other brain regions. Physiologically, inhibition of nociceptive neurons
and neurons with non-sensory discriminative response characteristics may be
involved in cognitive modulation and in the interaction of pain and touch (see
references in Schnitzler and Ploner 2000). This modulation may engage inhibitory
interneurons to quench local hyperactivity and/or synchrony, or even ongoing
inhibition (see Chapter 7). From a cellular point of view, it is important to note that
different classes of GABAergic neurons are not distributed homogeneously among
the different cortical layers (reviewed in Defelipe and Farinas 1992) and no less than
14 subclasses according to histologic and electrophysiologic (9 subclasses) criteria
exist in SI (Gupta et al. 2000). The majority displays depression at low frequencies,
but prominent facilitation at higher frequencies (50 Hz) and some show a burst
response at the onset of depolarization. Specific interneurons form specific types of
synapses on pyramidal cells and probably on other interneurons; while the temporal
dynamics of transmission of glutamatergic synapses is highly heterogeneous,
GABAergic interneurons form synapses with virtually identical temporal dynamics
onto different targets of the same class (GABA group). It is clear that the exact
mechanism of engagement of inhibition will require much more detailed work.
The tight coupling of sensation and motricity may also explain CS effects.
Suppression of natural pain-related behaviors clearly engages a potent volitional
motor control process, yet movements are known to increase the threshold for
detection and decrease the perceived intensity of somatosensory stimuli, including
those at a painful level (active movements having greater and more consistent effect

Neuromodulation

than passive movements). Humans perceive forces they exert as weaker than identical
forces acted upon them: in fact, a corollary discharge of the effort attenuates the
subjects sensory feedback (Shergill et al. 2003) and pain interferes with mental
representations of movement (Schwoebel et al. 2002). Tonic painful input leads to
inhibition of MI and SMA during motor performance on the painful side (and
contralateral one though less) (Binder et al. 2002). TMS studies show that under
normal conditions sensory afferents limit the activity of inhibitory neurons in MI,
and that after pure thalamic sensory stroke, MI intracortical inhibition is increased
(Liepert et al. 2005). In one possible scenario, the CP generator tonically inhibits MI,
but, if this is too intense, CS may not be able to engage inhibition itself.
Finally, a relatively high stimulation frequency can induce a tonic depolarization
and cortical inactivation effect, which is known to inhibit thalamic relays.
2. Deep brain stimulation (DBS) (Table 6.2)
In 1960, Mazars and colleagues first reported attempts to stimulate the somatosensory pathways, particularly the neospinothalamic tract at its termination in Vc,
for the treatment of chronic neurogenic pain. Their theoretical framework was the
theory of Head and Holmes, which held that CP might be the consequence of an
imbalance between protopathic and epicritic sensory functioning: stimulation of the
thalamic sensory relay nuclei would presumably increase the epicritic component and
hence inhibit the protopathic inflow (an anticipation of the later gate control theory).
Acute thalamic stimulation was later found to suppress the aversive behavior in
patients with facial postherpetic neuralgia (White and Sweet 1969). However, the real
interest in DBS for the treatment of chronic pain in humans arose in the 1970s.
Reynolds (1969) discovery that electrical stimulation of the rat midbrain could
produce profound analgesia without the concurrent administration of drugs and the
gate control theory (Melzack and Wall 1965), according to which stimulation of
large-diameter fibers is capable of inhibiting nociceptive information, paved the way
to most electrical stimulatory procedures. Despite initial optimistic reports, it soon
became clear that DBS was not as successful as was initially hoped. The clinical data
did not fit with animal findings, and large discrepancies were noted between the
results of different neurosurgical groups.
The targets for DBS include thalamic Vc nuclei and/or the posterior limb of the
internal capsule, the caudal medial thalamic areas around the third ventricle,
including CM-Pf and the junction of the third ventricle and the sylvian aqueduct
(rostral ventral PAG-caudal ventral PVG). CP is generally treated by contralateral
Vc stimulation, which is effective only unilaterally. The internal capsule (posterior
limb) may be used if thalamic tissue is unavailable (e.g., after an infarct or
encephalomalacia). A few groups also simultaneously stimulate the PVG area and Vc.
Mechanism of action.
a. PAG-PVG: Numerous observations made in patients, such as an increase
of the endorphin content in ventricular fluid after PAG-PVG stimulation, cross
tolerance between SPA- and narcotic-induced analgesia, and naloxone reversal of
PAG-PVG stimulation-induced suppression of chronic pain, support the notion that
pain relief by PAG-PVG stimulation is mediated by endorphin-containing neuronal

199

200
Paraplegic pain (5 pts, then 19)

Richardson and Akil (1977)

Thalamic CP (28 pts)

SCI (8 pts)
CP (thalamic pain: 2 pts; partial SCI
pain: 3 pts; postcordotomy CP: 1 pt)

BCP (incl. CPSP) (5 pts)

SCI (5 pts)

CPSP (thalamic) (1 pt)

CCP (iatrogenic) (2 pts)
Thalamic pain (1 pt)
SCI pain (1 pt)
SCI pain (2 pts)

Lazorthes (1979)

Schvarcz (1980)

Mundinger and Salomao (1980)

Mundinger and Neumuller (1982)

Ray and Burton (1980)

Plotkin (1982)

Richardson et al. (1980)

Vc (bilat. in SCI) or IC PAG/PVG

Thalamic lesion (3 pts)

Brainstem lesion (6 pts)
SCI (4 pts)
BCP/CCP

Mazars (1976)
Mazars et al. (1979)
Includes all previous papers by this
pioneer group on the topic

Vc
Vc
PVG

CM-Pf

IC/LM (3 pts)
Pulvinar (1 pt)
PAG/PVG (1 pt)

IC/ML (4 pts)
Pulvinar (1 pt)

Medial posteroinferior thalamic areas

Vc

PAG-PVG

Target

Type of pain

Author(s)

TABLE 6.2. Deep brain stimulation (DBS) arranged by reporting groups

0% (?)
0% (?)
0% (?) successes (follow-up: 642 mos)

450% relief in all, drugs not stopped, effect

abates in time

0%, 50% and 5070%

470%
50%
(except one, follow-up shorter than 2 yrs)

470%: 1; 5070%: 1; 50%: 3 (1 pulvinar)

(max. follow-up: less than 2 yrs). No relief at
longer term.

Pain relief (deep background pain and hyperpathia): 475% (but never 100%) relief: 2
5075% relief: 2
Failure: 2 pts
Hyperpathia abolished, deep background pain
only reduced. No reversal by naloxone.
Follow-up: 642 mos

Successful pain relief in 5

Successful pain relief in 2

Significant pain relief in 2 (18 mos). 1 pt

previously submitted to failed rhizotomy/
cordotomy.
Further series: good relief at 1 yr in 6 pts

First group to stimulate the thalamus starting

1960

Failure
Relief in 5
Relief in 4
Poor results

Results/notes

201

IC (8)
IC Vc (1)
IC Vc ML (1)

Vc
Vc

PAG
PAG
Vc

CP (thalamic and putaminal stroke:

9 pts; extrathalamic subcortical: 1 pt;
MS CP: 1 pt)

SCI pain (1 pt)

CP above brainstem (8 pts)

Myelopathic CP

Namba et al. (1984)

Frank et al. (1984)

Tsubokawa et al. (1985)

Katayama et al. (2001)
Includes all CP patients submitted to
DBS by Tsubokawas group

Heiss et al. (1986)

Thalamic CPSP

19701984

Postcordotomy CP (9 pts)

BCP (cortex, thalamus, brainstem)

(13 pts)
Paraplegia CP (8 pts)

At discharge: 100% (3), 5095% (3), fair

(drugs needed, 2), 0% (3; 1 with thalamotomy,
pulvinotomy, mesencephalotomy). Best
stimulating point for analgesia not in the center
of posterior limb but in most posteromedial
part (area triangularis)

Vc

CP (including SCI)

Turnbull (1984) Also includes Shulman

et al. (1982) and other previous papers
by this author

Hosobuchi (1986)
Includes all previous published pts

Of limited efficacy, particularly ineffective in SCI

pain. One pt with brainstem stroke relieved
over a few years

Vc

CPSP, thalamic (2 pts)

Vc (likely, not specified)

Vc, lemniscal, PAG

(continued)

Pain relief (follow-up: unavailable)

8 early successes, 5 failures; 6 late successes,

2 failures
3 early successes, 5 failures; 2 late successes,
1 failure
8 early and late successes (75100% relief); 1
early bleeding
PAG DBS: ineffective; lemniscal: 36% success
Follow-up: 214 yrs

Short-term relief: 80% in 2/8 pts; 6080% in

3/8 pts; <60% in 3/8 pts. Long-term relief:
33%
No relief
No relief
6080% relief in 2

Poor result

4060% pain relief (1218 mos)

Good or excellent results (follow-up: up to 18

mos)

Broggi et al. (1984)

Right CM-Pf and left CM stimulation

CPSP (2 pts)

5 slight late reliefs (6 mos4.5 yrs)

Andy (1983)

PVG/Vc

CP, thalamic (5 pts)

Dieckmann and Witzmann (1982)

202
Vc
Vc/IC

Case 1: thalamic stroke with left pain; 6

mos later, left stroke with right pain
CP (12 pts)

Tasker et al. (1991, 1992)

Includes all published cases from
Toronto Western Hospital

PVG
Vc

Crisologo et al. (1991)

Partial SCI pain (17 cases)

19731989

Thalamic CP (19 pts)

Relief in 5 (3 with evoked pain: 2 relieved),

failure in 7 (6 with evoked pain: stimulation
painful in 3)

Insignificant relief

Long-term: 5 very good, 7 good, 3 fair, 4 poor.

Better results in parathalamic lesions than true
thalamic lesions
Pain relief in 3
5 very good, 8 good, 3 fair, 1 poor
DBS for MS CP: effect lost in time

PAG/PVG (both in 3)
PAG/PVG

CP, thalamic (3 pts)

Paraplegia pain (7 pts)

Vc

Vc or PAG/PVG

SCI CP

Siegfried (1991)
Includes all previously published
personal cases

Test stimulations: 14 VPL, 11 VPM, 6 IC. Pain

relief sufficient for internalization in VPL:
9/14 pts (64%); in VPM: 9/11 pts (82%); in
IC:
1/6 pts. Initial success rate: 56%; long-term
pain relief: 24%.
14 electrodes implanted (7 Vc, 7 PAG-PVG) in
11 SCI pts. Pain relief sufficient for
internalization in 2/11 pts (18%).
No persistent (46 wks) pain relief.
Unsatisfactory pain relief, no internalization
7 electrodes implanted; 2 internalizations; no
persistent pain relief (0%).
6 Vc and 2 PAG-PVG electrodes implanted;
2 Vc and 1 PAG-PVG electrodes internalized.
3/5 pts (60%) with initial successful
stimulation, 2/5 (40%) long-term pain relief.
Follow-up: 24168 mos; paresthesias independent of analgesia, not vice versa.
CP relief approaches 30% (rate close to that
expected from placebo)

Vc or IC

CP (25 pts)

Levy et al. (1987)

Includes Fields and Adams (1974),
Adams (19771978)

Postcordotomy CP (5 pts)

Results/notes

Target

Type of pain

Author(s)

TABLE 6.2 (continued)

203

19781993

Vc /
PAG/PVG

PAG-PVG

CP, thalamic. Failure

Unilat. PAG Koelliker-Fuse nucleus

(1 pt)
PVG Koelliker-Fuse nucleus (2 pts)

BCP (14 pts)

CCP (12 pts)

Young et al. (1995)

Includes all pts appearing in
previous publications

(continued)

Excellent pain relief in 2 pts suffering from SCI

CP (follow-up 2 yrs and 8 mos, respectively). In
1 pt cessation of stimulation after 2 yrs was not
followed by a full-fledged return of pain.
Additive effect from PVG-Koelliker-Fuse n.
simultaneous stimulation (but KF 4 PVG)
Excellent or good pain relief from PAG-PVG DBS
only in 35% of pts (median follow-up 4 7 yrs)
(From previous series.) Excellent pain relief
(Vc): 1; partial relief (Vc PAG-PGV): 9;
ineffective: 6
(Of SCI pts, 4 had 50% relief at 260 mos)
Apparently unsatisfactory long-term results
from PVG stimulation in CCP.
Analgesia onset: within minutes; long
after-effect in some pts

4/6 reliefs; follow-up: 16 mos

Centrum medianum

CP, thalamic (6 pts)

Hariz and Bergenheim (1995)

Vc

Thalamic pain (5 pts)

SCI pain (5 pts)
Postcordotomy CP (1 pt)

Vc (mostly bilat.)

PVG either ineffective or inferior to thalamic

stimulation with the exception of 1 CCP pt
whose severe allodynia and hyperpathia
disappeared acutely after 510 min of PVG
stimulation
Steady pain relief 450%: 20% of pts;
2550%: 16% of pts;
Intermittent pain relief: 0%;
Evoked pain relief 2550%: 16% of pts
Global: relief in 3
3/5 pts initial pain relief; 1/5 long-term
benefit
Short-term pain relief in 3/5; long-term pain
relief in 2/5 pts
Failure

Gybels et al. (1993)

CCP (13 pts) (complete lesion or

incomplete lesion unresponsive to SCS)

PVG

204
CP (brainstem or suprathalamic origin)
(6 pts)
SCI (3 pts)
(Eur. Coop. Study)
19851997
CPSP (1 pt)
CPSP (thalamic stroke) (1 pt)
CPSP (14 pts) (1 pt) (cortical: 5;
thalamic: 8; pontine: 1; IC: 1)
Other CPs (5 pts)

Blond et al. (2000)

Phillips and Bhakta (2000)

Krauss et al. (2001)

Nandi and Aziz (2004)

Owen et al. (2006)

19952005

CPSP (1 pt)

Improvement
Failure
In 1 patient, trial PVG DBS provided 0% relief.
12 patients seen for an average of 16 mos
(336 mos). One patient had less than 3 mos
follow-up. 11 of 14 were satisfactorily relieved
and opted for IPG. 13 of 19 consecutive CP
patients had satisfactory control with PVG and/
or Vc DBS. Trial relief maintained over an
average 16 mos in all but 2 pts. Vc stimulation
alone reasonably suppressed the pain in 4 pts
(MS, tractotomy, post-SAH stroke, Chiari);
however, in the first 2, paresthesias were
intolerable. In the other 2 PVG DBS alone was

CM-Pf Vc
Vc PVG (16 pts)
PVG (1 pt)
Vc (1 pt)

Pain relief 450%: 1/3 pts

Unsatisfactory results. Paroxysmal pain

refractory

Partial relief (analgesic reduction) of

spontaneous and evoked pain. MCS
ineffective. Painful relapse after tumoral
displacement of electrode

Short- and long-term (3.4 yrs) successful

(5075%) pain relief in 1; early failures
(050% pain relief) in 4
Early successful pain relief (51100%),
1; early failures (050% pain relief), 2;
late failures (2 yrs), 3
Analgesia within 10 min (bipolar stim.);
duration of pain pre-DBS not prognostic

Results/notes

PVG

Vc DBS

Vc DBS

Vc

SCI pain (3 pts)

Barraquer-Bordas et al. (1999)

Vc (1) IC (4)

CPSP (thalamic) (5 pts)

Kumar et al. (1997)

Includes all pts from 1990 paper

Target

Type of pain

Author(s)

TABLE 6.2 (continued)

205

Romanelli and Heit (2004)

CPSP (1 pt)

Vc DBS

100% relief over 455 mos with several

changes of parameters

superior. Combined Vc-PVG DBS was never

synergic and worsened the pain in 2 pts.
Their Fig. 2 with results on 14 pts (2 pts not
shown having less than 3 mos follow-up): 3 pts
not implanted (2 having less than 10% relief
but 1 40%: why not implanted?). In 7 relief at
follow-up was slightly better than test relief, but
in 4 it was less, in 1 case half of it; never
100% relief or somewhat less. Final series of
CPSP pts only (2006): 15 pts, evaluated with
VAS, MGPQ, PRI-R. Pts with Vc strokes only
implanted in PVG-PAG; avrg. follow-up:
27 mos, but results plotted at 2 yrs; mean
relief (VAS) for cortical strokes: 42% for all
others: 54%; opposite results with PRI-R (!).
Wide range of improvements: from slight
worsening to 91.3% improvement. 7 pts
stopped all analgesics.
Post-effect: for over 24 hours
Severe burning hyperesthesia most responsive.
Most pts preferred PVG DBS to Vc DBS
(results thus refer mostly to PVG DBS)
Once burning abates, pts note the background
crushing, aching sensation more strongly (past
authors may have exchanged this phenomenon
for tolerance and relapse)

206

Central Pain Syndrome

systems (see references in Gybels and Kupers 1995; Meyerson and Linderoth 2001).
However, this hypothesis has been firmly challenged by Young and Chambi (1987).
Using a double-blind, placebo-controlled study design, they found no evidence that
PAG-PVG-induced SPA in humans is mediated by an opioid mechanism. In a study,
low- (120 Hz) and high-frequency (50 Hz) stimulation of the PAG produced
neither relief nor reproduced pain in 8 thalamic CPSP, 1 tumor thalamic CP, 1 SCI
pain and 1 tabes dorsalis patients, despite a modest-to-significant increase of CSF
endorphin levels (Amano et al. 1982): this increase was interpreted as a psychological
response.
The Oxford group (Nandi et al. 2003; Nandi and Aziz 2004) found that pain
suppression is frequency-dependent. During 535 Hz PVG stimulation, the
amplitude of thalamic field potentials (FPs) was significantly reduced and this was
associated with pain relief; at higher frequencies (50100 Hz), there was no reduction
in the FPs and pain was made worse. Switching on the stimulation was followed
immediately by a change in the thalamic potentials; however, the FPs did not revert
to baseline immediately on cessation of stimulation, but only after a lag of 515
minutes depending on the duration of stimulation. The FPs consisted of a very low
frequency potential, of 0.20.4 Hz, in Vc: their amplitude seemed to correlate with
pain intensity, being much stronger off or with 50 Hz stimulation when there was no
pain suppression, than with 535 Hz stimulation with accompanying pain relief. This
suggested a fairly direct neuronal circuit between PVG and Vc mediated by
reticulospinal neurons. All patients were also stimulated in Vc, alone or simultaneously with PVG. The PVG FPs were independent of both the pain scores and the
state of stimulation of Vc. In non-responders, there was no flattening in the slow
wave thalamic FPs across different frequencies of PVG stimulation.
b. V C : The mechanism by which Vc DBS works is not likely to result from the
activation of an endogenous opioid system (or other descending fiber tracts), because
its analgesic effect is not reversed by naloxone. Although investigators found that,
after thalamic stimulation, beta-endorphin levels were more than twice the resting
level, no differences in beta-endorphin levels could be demonstrated between
patients reporting complete pain relief and those reporting only partial relief
(Tsubokawa et al. 1984); a much higher increase in beta-endorphin levels was found
after PAG stimulation. In humans, administration of an antidopaminergic agent
antagonized the analgesic effect of Vc, but not PAG, DBS (Hosobuchi 1990).
However, Velasco and colleagues (1998) found that acute CM-PF stimulation at
60 Hz elicited pain in epileptic patients and this reaction was blocked by an opioid
agonist. Tsubokawa and Moriyasu (1975) found that Vc DBS in 2 of 4 pain (non CP)
patients at 50100 Hz inhibited Center Median nociceptive responsive thalamic
neurons, while stopping the pain.
A direct action on spinal STT neurons has been excluded, even via relay in the
brainstem (Vilela Filho and Tasker 1994); also in light of results of drug dissection of
CP, it is not clear if NE, 5HT or other fibers/nuclei are involved, and to what degree
(see discussion in Gybels and Kupers 1995; Meyerson and Linderoth 2001). Andy
(1983) suggested that altering the excitability state and/or the thalamic discharge
patterns by artificially induced electrical stimulation underlie the pain-relieving

Neuromodulation

effects of DBS, i.e., jamming a low threshold discharging pain system (Emmers).
However, an exclusive depolarization block is an unlikely explanation. A differential
effect on both gray (inhibition) and white matter (excitation) should also be
considered (Bejjani et al. 2002). Electrical stimulation of the neuropil in general
affects axons rather than cell bodies, thick before thin myelinated axons and
preferentially fibers parallel to the stimulating current more than those transversally
(Nowak and Bullier 1998; Ranck 1975). Neural elements up to 25 mm from the
stimulating cathode may be excited (Ranck 1975). Presently, a GABA release is
considered a likely mechanism (e.g., Obeso et al. 2000), although inhibition may
adapt with continuous stimulation (Ashby and Rothwell 2000). A possibility would
be orthodromic stimulation of inhibitory afferents to a target structure or recruitment of local inhibitory interneurons (Ashby and Rothwell 2000). Vc stimulation
too can suppress medial thalamic hyperactivity (Tasker et al. 1983). Since the
thalamocortical loop probably works more like a nonlinear dynamic system that is
not solely based on a firing-rate code, DBS may actually work by rebalancing
a skewed oscillatory pattern (Canavero 1994).
Neurometabolic studies have been published on this problem. These studies
reported stimulator-induced signal increases to be higher than task activations
(maximum 2%). Heiss and colleagues (1986) studied one CPSP case with PET.
At rest (pain condition), the lowest metabolic rate was in the infarcted thalamus;
some areas showed decreased glucose consumption in the otherwise normal
ipsilateral cortex. A second PET during DBS (off-pain condition) revealed markedly
decreased glucose metabolism in most brain regions. Rezai and colleagues (1999)
scanned (fMRI) two patients who had steady-burning CP due to traumatic SCI
(a third had PNP). PVG DBS in contrast to Vc DBS did not activate SI, but the
cingulate cortex (compare to Vim DBS for tremor). Low-frequency stimulation of
PVG led to activation of the medial thalamus (compare with Nandi et al. 2003).
Activations near the electrode were written up to a possible, local nonspecific CBF
increase rather than neural pathway activation. At paresthesia-evoking intensities, Vc
DBS resulted in the activation of SI in all 3 pain patients. In most cases, areas of
cortical activation corresponded to the homuncular somatotopy of paresthesias (3 V,
75100 Hz, 150200 ms). With no paresthesias, SI was not activated. In addition to
SI, there was activation of thalamus, SII and insula. In a similar study, Duncan and
colleagues (1998) submitted 5 patients with neuropathic pain (perhaps inclusive of
CP) to Vc DBS. All had obtained relief for more than 3 years to reduce a placebo
confounding role. Three patients were relieved, while two had no immediate relief.
They reported that <100 Hz Vc DBS increased rCBF in and near the thalamus and
some cortical areas, the effect being more prominent with continued stimulation.
Their data did not support activation of tactile thalamocortical pathways being the
sole mechanism underlying successful Vc DBS. Their most prominent cortical rCBF
increase was in ipsilateral anterior insula, both with relief and not, although
somewhat stronger with relief. Patients perceived both paresthesiae and cold and
warmth during stimulation. The close proximity of microstimulation sites evoking
tactile and thermal sensations indicates that bipolar stimulating electrodes could
easily stimulate neurons within both the insular and SI pathways. They also observed
a nonsignificant trend toward activation in ACC with Vc stimulation. Davis and

207

208

Central Pain Syndrome

colleagues (2000) studied two patients with CCP (plus 3 other neuropathic pain
cases) submitted to Vc/ML stimulation. The first was a paraplegic suffering from
unilateral leg pain: he obtained 100% relief after 30 minutes of stimulation. This
analgesia disappeared immediately upon cessation of DBS. Follow-up was 9 months.
On PET day, he was on amitriptyline, baclofen, diazepam and oxycodone. The
second suffered from spinal AVM-related CP to left leg. Follow-up was 16 months.
Analgesics were retained for 12 hours before PET. There was 0% relief at follow-up,
but some relief immediately postoperatively (thalamotomic effect?). Paresthesias
were strongest at the beginning of stimulation and subsided as stimulation continued. There was no clear relationship between the degree of stimulation-evoked
pain relief and the magnitude of rCBF change in either region of the ACC (BA3224).
Activation of posterior ACC was detected after 30 minutes of DBS, but not at the
onset of stimulation, in contrast to the ACC, which was activated throughout the
period of DBS. Thus, posterior ACC was not related to direct activation from
thalamus, but to other structures. Duncan and colleagues (1998) also noted that
some of their DBS-induced activations were stronger after 30 minutes of DBS than at
DBS onset; unlike this study, patients in Davis study did not experience thermal
sensations during DBS and no insula activation was seen. Lack of activation of SI-SII
could be explained by low statistical power (only 2 responders), paresthesias in
different body regions, thus activating different portions of SI-II, or diminishing
paresthesias in the course of DBS. Other CBF changes may have involved other
cortical and subcortical areas.
c. O T H E R A R E A S : Septum and caudate nucleus stimulation has never been reported
in CP patients. Basal ganglia, known to process noxious information, and medial
thalamic nuclei (Haber and Gdowski 2004) are closely interconnected, but
stimulation at these levels is not expected to relieve CP.
Mayanagi and Sano (1998) state that patients with chronic pain of thalamic or
spinal origin failed to experience pain relief with hypothalamic DBS-like stimulation. Failure of stimulation to relieve CP may be similar to generally ineffective
results of PAG-PVG DBS.
Stimulation of the Koelliker-Fuse nucleus, a pontine satellite of the locus coeruleus
and the major source of catecholamine-containing fibers to the spinal cord, has been
attempted in CP cases, but patients were too few for meaningful conclusions.
Other areas of possible interest, but not yet clinically explored, include the anterior
pretectal nucleus.
Efficacy. Results of DBS for CP have not lived up to expectations, providing no
long-term benefit, but in a few cases in most series.
Whereas patients referred for DBS are those in whom the success rate of many
prior therapies has been zero or close to it, this notwithstanding, long-term results
remain unsatisfactory.
According to Gybels and Kupers (1995), in their review up to 1993, results of
DBS are as follows.

Neuromodulation

a. V C D B S : Thalamic pain. Of 100 reported cases, mean success rate was 36%. The
median success rate based on seven studies was 30% (range 063%).
SCI pain (mostly paraplegia pain). In 63 patients, mean success score was 35%.
The median success rate based on eight studies was 25% (range 0100%).
Postcordotomy pain. Of 26 patients, 19 (73%) responded well to DBS. Median
success rate of four studies was 84.5% (range 40100%).
b. P A G - P VG D B S : These are generally poor for thalamic and paraplegic pain, with
some exceptions.
First, there can be a large placebo effect (Marchand et al. 2003). Secondly, CP
includes several components (Tasker 2001) which may be differentially responsive to
stimulation. Moreover, CP fluctuates, and a few successes may simply be due to
a spontaneous downward fluctuation.
In the longest followed-up series of Vc-IC DBS (Levy et al. 1987), CP relief
approached 30%, a rate close to the level expected of a placebo effect. They reported
0% relief for paraplegia pain. All patients with thalamic pain had a long-term success
rate of 24%; the success rate of Vc DBS was about 43%, and if cases in whom Vc
DBS produced paresthesias only are considered, the long-term success rate was
about 55%. Long-term relief was obtained in 40% of postcordotomy pain. Gybels
and Kupers (1995) stress the fact that true relief may be lower than suggested by
the literature.
It is interesting to compare some series of DBS for CP with regard to long-term
successes. Those with a short follow-up boast successes in the 7080% range, while
the longest followed-up series (Levy et al. 1987: mean, 80 months) reported longterm relief in about 30% of patients. The lesson is clear: pain relief abates with time
(see also discussion in Nandi and Aziz 2004; Owen et al. 2006). So-called tolerance
to DBS, despite initial claims, has not been reversed by disulfiram, L-tryptophan,
amitriptyline, temporary holidays, while alterations in stimulus parameters have
sometimes proved effective (Young and Rinaldi 1997): this is not due to tissue or
endorphin changes (Tsubokawa et al. 1984), but electrophysiological adaptations.
Young (1995) originally believed that most patients who experience declining
effectiveness of DBS did so over the first year, but long-term follow-up over more
than 15 years indicates a steady decline: total pain relief remained possible only in
a few cases. According to Kumar and colleagues (1990), there is an initial two-year
fall-off of pain control caused by idiopathic tolerance, with stable results thereafter,
regardless of site of implant, suggestive of some biochemical modification of tissues
around the electrode. Romanelli and Heit (2004) suggested that changing parameters
at the first hint of relapse may block tolerance and restore relief; however, relief can
be lost suddenly without warning.
According to the thoughtful review of Duncan and colleagues (1991), (1) the
majority of the clinical reports are case histories rather than well-controlled studies,
(2) the pain measures described usually involve imprecise questions about pain relief
that do not allow a rigorous statistical evaluation, and (3) studies are rarely
conducted in a double-blind fashion, and data from placebo-controlled experiments
are seldom included. The potential for at least short-term placebo responses

209

210

Central Pain Syndrome

is substantial, considering the elaborate nature of the surgical procedure, the

mysterious electronic technology involved and the close interpersonal relationship
that develops between the pain patient and the attending clinician. No study
provided a statistical analysis of the clinical pain changes. The absence of wellcontrolled studies and statistically significant results prohibits an objective appraisal
of the clinical efficacy of DBS. In fact, there appears to be an astounding variability in
reported results from several centers. It is improbable that this variability can be
accounted for by differences in pain pathology because (1) in the larger studies, the
major pain syndromes are all approximately equally well represented, and (2) even
when the results obtained in a particular diagnostic category are compared, the same
variability between the authors remains. The larger and older series generally
reported much more favorable results than did the smaller and more recent series.
The data of Marchand and colleagues (2003) suggest that for some patients DBS can
be helpful in reducing clinical pain, but effect is moderate, as with SCS (see below).
Importantly, patients reported the presence of paresthesias even in placebo conditions (the ability to induce paresthesia in the painful area is considered important
for target localization!). Patients expectations are an important factor in the DBS
placebo effect. DBS but not placebo DBS was found to produce a significant
reduction in thermal noxious (but not tactile) perception. The conclusion was that
a strong placebo effect may be involved in the efficacy of any form of DBS and placebo
effects can last even for up to 5 years. Interestingly, Wolksee and colleagues (1982)
found no statistically meaningful difference between Vc and sham stimulation.
The PAGPVG region responsible for analgesia is small (Gybels and Sweet 1989;
Duncan et al. 1991), and also thalamic size varies considerably from patient to patient
(Young 1989); thus, extreme precision is needed for deep stimulations, otherwise
results will be jeopardized. Stereotaxic atlases are only a starting point and MRI
and microrecordings are employed for fine positioning. Several factors have been
proposed to influence therapeutic outcome and hence account for the observed
variability between different authors. However, most of these explanations are based
on empirical observations, and they have not been confirmed in controlled studies.
Among these are stimulation parameters and electrode configuration (520 Hz
versus 30100 Hz; stimulation intensity below or above the level at which
paresthesias are felt; brief periods of stimulation versus longer periods; dissociation
between pairs of contacts producing analgesia versus paresthesias), exact target
localization, patient selection (e.g., long-term success is largely reduced in a hysterical
personality or a patient with secondary sickness gain), pain type (steady versus
evoked). Since physiological studies of Vc stimulation indicate that strong inhibition
of nociceptive neurons occurs at frequencies higher than those frequently used
clinically, it may be that human stimulation parameters have not been systematically
optimized (Duncan et al. 1991). The review of Gybels and Kupers (1995) found that
not all authors reported early treatment failures (i.e., failure during test stimulation),
and hence results overestimate the real therapeutic efficacy. Decrease in success rate
occurred both in patients with PAG-PVG stimulation and in patients with Vc
stimulation. Finally, many authors use as a criterion of success a pain relief of 50%
or more, implying that several patients continue to be unrelieved of their pain: even
moderate pain may be crippling and only a few patients obtain total relief over

Neuromodulation

several years. Taskers group (Vilela Filho 1996) reported 6 BCP (not brainstem) with
evoked pain complaining of unpleasant paresthesias with Vc-ML-IC DBS. They all
had unpleasant paresthesias with previous SCS, restricted thalamic lesions on CT and
associated intermittent pain. Minor risk factors were cold allodynia-hyperpathia
and no sensory loss.
Most importantly, DBS has complications which can be lethal. Bleeding is
associated with a mean mortality of 0.3% and a permanent disabling morbidity rate
of 1.4% (Favre et al. 2002). The risk of bleeding from DBS is related more to the
patient (vascular fragility, various coagulopathies, unstable blood pressure) than to
the type of stereoprocedure performed. Young and Rinaldi (1997) in 178 patients had
3.9% permanent complications and 0.6% indirect deaths. Higher mortality was
reported in older series of DBS (<1.6%) (Bendok and Levy 1998).
3. Spinal cord stimulation (Table 6.3)
The gate control theory of pain (Melzack and Wall 1965) inspired Shealy to implant
the first dorsal column stimulator in a cancer patient, the dorsal columns being
rich in the large, low-threshold A-beta fibers, alleged to close the gate against
nociception-subserving afferents and also led to peripheral nerve stimulation.
However, pain reduction without paresthesias can be obtained also from electrodes
placed over the anterior cord surface (references in Gybels and Sweet 1989).
Mechanism of action. This is basically unknown. Certainly, SCS does not activate
any gating mechanism, or it would also block acute pain. SCS may modulate local
spinal networks, but also thalamocortical areas: the amplitude of evoked potentials in
the human somatosensory cortex (Larson et al. 1974) and thalamic CeM nucleus
(Nyqvist and Greenhoot 1973) is reduced by SCS; SCS also reduced the firing rate
(including bursting) of thalamic Centrum Medianum neurons, with a poststimulation effect of a few hours, at parameters achieving partial relief, in a patient
with mixed nociceptiveneuropathiccentral pain (Modesti and Waszak 1975).
Taskers group (Kiriakopoulos et al. 1997) reported on a SCI pain patient who
described paresthesias and relief of her left leg pain at 2V, but not 1V: fMR showed
increased activity in the right sensory cortex at 2V compared to 1V stimulation.
SCS may modulate several transmitters and peptides (5HT, glycine, adenosine),
but the evidence favors GABA; in consideration of the efficacy of different GABA
agonists, a role for both GABA A and B receptors can be envisioned, with an action
on WDR cells (also see Meyerson and Linderoth 1999). If SCS acts by engaging
GABA neurons, some may have died following excitoxic post-trauma injury (human
studies show an increase of glutamate in such situations; see Canavero et al. 2003),
and are no more available.
Efficacy. A prerequisite for successful pain relief by SCS has usually been a coinciding or blanketing of the painful area by the generalized paresthesias, but evoked
paresthesias do not guarantee pain relief, and evoked sensations can also be outside
the painful area.
Marchand and colleagues (1991) provided the first placebo-controlled study of
SCS for chronic pain (other than CP). The conclusion was clear-cut: reduction in

211

212

Central Pain Syndrome

TABLE 6.3. Spinal cord stimulation (dorsal column stimulation) (SCS)

Author(s)

Type of pain

Results/notes

Nashold and Friedman (1972)

SCI pain (leg pain; 6 pts)

Excellent: 1/6 pts (follow-up:

11 yrs)
Partial: 4/6 pts (mild analgesic still
required)
Unsatisfactory: 1/6 pts

Nashold (1975)

CPSP (3 pts)

Initial pain reduction with

stimulation of the trigeminal tract in
the upper cervical cord

Urban and Nashold (1978)

CCP (3 pts)

Pain relief: 1; unsuccessful test

stimulation (no paresthesias): 1;
lost to follow-up but initial pain
relief: 1

Sweet and Wepsic (1974, 1975)

Postcordotomy dysesthesia (7 pts)

MS (3 pts)
SCI pain (4 pts)
Myelopathic pain (7 pts)

Good relief: 2
Good relief: 1
Failure
Failure
Hyperpathia never relieved

Hunt et al. (1975)

Radiation myelitis CP (1 pt)

0%

Long and Erickson (1975)

SCI CP (1 pt)
Postcordotomy CP (2 pts)

Failure
Failure

Lindblom and Meyerson (1975)

SCI pain (2 pts)

1 early success

Sedan and Lazorthes (1978)

Cord CP (postcordotomy pain:

14 pts; SCI: 16 pts)

Rosen (1979)

MS

Postcordotomy pain: review of

Sweet, Shelden, Nashold and Long
reports (14 pts). SCS results:
excellent: 3/14 pts; bad: 1/14 pts;
failure: 10/14 pts.
SCI pain: review of Sweet and Long
reports (16 pts).
SCS results: excellent: 1/16 pts;
fair: 2/16 pts; failure: 13/16 pts (at
least 1 pt with above-lesion SCS).
No screening test in any pt. BCP in
anybodys experience: SCS totally
ineffective
Good relief in 20%, 0% in 60%
of pts

Richardson et al. (1980)

Paraplegia pain (10 pts)

SCS rostrad to lesion. Pain relief

450% from test stimulation:
5 (3 with incomplete cord lesion).
At 1 yr follow-up: 4/5 lost to
follow-up (2 pts died, 1 lost after
3 mos); 1/5 pain relief (presumably
from recovered lesion)
Failure of test stimulation in 5 pts
(3 with complete cord lesion)

Neuromodulation

213

Author(s)

Type of pain

Results/notes

Demirel et al. (1984)

CP (10 pts)

Positive trial test in 6/10 pts.

No late results

Vogel et al. (1986)

CP (3 pts)

No response to trial stimulation

in all

Wester (1987)

MS CP (3 pts), SCI CCP (3 pts),

tumor CCP (1 pt)

Benefit at 15 mos (median; range:

460 mos): 0% MS CP, 33% SCI
CCP, 0% tumor CCP.
Comment: global effect restricted,
dwindling effect in time, DCS not of
any great help

Mittal et al. (1987)

CP (8 pts)

Positive trial test in 3 pts. Persistent

pain relief (3 mos, 8 yrs): 2 pts

Beric et al. (1988)

CP

SCS may worsen CP with absent STT

function and preserved DCs

Buchhass et al. (1989)

SCI pain (7 pts)

6/7 good/very good relief at

372 mos

Krainick and Thoeden (1989)

CCP (transverse spinal

lesions: 2 pts, other spinal injuries:
2 pts; incomplete conus-cauda
lesion: 4 pts; tetraspasticity after
cervical disc operation: 2 pts)

Initial pain relief in all pts; no

long-term follow-up
Overall (CP plus other pains)
long-term (23 yrs) results:
5075% pain reduction in 39% of
pts.
60% had complications requiring
removal of the stimulator

Michel et al. (1990)

CPSP (5 pts; parietal)

50% pain relief in 2

Cole et al. (1987, 1991)

CCP (4 pts)

0% (1 worsened)

Simpson et al. (1991)

Thalamic CP (9 pts)

3 significant, 3 modest, 2 no
benefit, 1 worsened (one after initial
modest benefit)
Worsened
6 complete/partial, 1 nonsubstantial, 2 failures (1 worsened)
(Relief: significant [complete or
partial pain relief, with significant
effect on medication and life-style,
praise of the apparatus by the
patient], modest [no substantial
benefit, no significant change in
medication, activity, sleep pattern],
failure)
Long-term follow-up data not
available for single disease. Median
overall follow-up: 29 mos
(2 wks9 yrs)

Post-thalamotomy CP (1 pt)
Painful paraparesis, paraplegia and
hemiparesis (10 pts)

(continued)

214

Central Pain Syndrome

TABLE 6.3 (continued)

Author(s)

Type of pain

Results/notes

Simpson (1999)

1 new CP, thalamic

Worsened.
Conclusion: SCS relief very unlikely
in complete SCI and reasonably
likely in partial SCI; unlikely in BCP

Spiegelmann and Friedman (1991)

Cord CP: SCI, MS (6 pts)

Positive stimulation test: 4 pts.

Long-lasting 50100% pain relief:
3 pts. Mean follow-up: 13 mos
(330 mos). No further pain relief
after a change in the distribution of
paresthesias in 1 SCI pain pt
(initial 1 yr benefit). TENS was not
predictive (TENS failures could
respond to SCS, as found by many
other groups)

Ohta et al. (1992)

SCI pain (4 pts)

At 1 wk, 100% relief in all. However,

at 35 mos, no relief in 3, while in
the fourth 7080% relief at 19 mos
only when SCS turned on

Tasker et al. (1992)

Taskers group

SCI complete (11 pts)

Steady (burning or not) pain unrelieved in 80% of pts. 2550% relief

in 20% of pts. Intermittent or evoked
pain unrelieved in 100% of pts. All
cases drawing benefit had T10L2
lesions.
(22/24 implants): steady pain relief
50% in 27% of pts and 2550%
in 14% of pts. Intermittent pain
unrelieved. 2550% evoked pain
relief in 25% of pts. Of cases
relieved, two thirds had T10L2
lesions.
Authors conclusions: SCS is more
effective for relief of steady pain
(36%) than of intermittent (0%) or
evoked pain (16%) (statistically
significant difference). SCS is
ineffective even for steady pain in
cases with complete lesions (20%
relief) Follow-up: 41 yr
Failures usually associated with an
inability to induce paresthesias in
the area of pain, due to severe cord
lesions inducing dorsal column
atrophy (dieback), difficulty in
accessing the epidural space

Incomplete (24 pts)

Neuromodulation
Author(s)

215
Type of pain

Results/notes
(trauma or previous surgery),
difficulty in producing paresthesias
over the large area of patients pain.
Failures not due to intrinsic
resistance of CCP to SCS.
Pain relief 450% for 1 yr only in 1
Positive stimulation trial: 7 pts; test
worsened pain in 2 pts with evoked
pain (just like Vc DBS in BCP pts
with allodynia). Early failures (pain
relief <50% within 1 yr of
implantation): 2/7 pts (early
success probably a placebo effect);
late failures (past 1 yr): 3/7 pts
Long-lasting (mean follow-up:
3.9 yrs, range 0.39 yrs) 450%
pain relief: 2/7 pts.
Drug reduction not specified nor
enhanced ability to work
Permanent implants in 90% of
cases. No detailed follow-up
reported

Kim et al. (2001)

(Taskers group)

BCP 12 pts
CCP 20 pts

North et al. (1993)

SCI pain (11 pts)

19721990

Italian cooperative study

(Broggi et al. 1994)

Paraplegia pain (23 pts)

Failure in all implanted pts within 1

yr of surgery, despite initial benefit
in several in this highly select group

Cioni et al. (1995)

Includes all previously published
cases of Meglios group in Rome
(PACE 1989; 12:70912,
J Neurosurg 1989; 70:51924)

SCI pain (25 pts)

Pain due to trauma or surgery at all

spine levels. 75% relief at the end
of the test period: 40.1% of pts.
Patients with more than 50% pain
relief at a mean follow-up of 37.2
mos: 18.2%. Better results in
patients with painful spasms and
constrictive pain in the transitional
zone in pts with incomplete thoracic
lesions. Below-level burning pain
unrelieved.
Authors conclusions: the relative
integrity of the dorsal column is an
important prerequisite for analgesia.
0% benefit without paresthesias
evoked in the painful area.
SCS not effective in treating true
SCI CP
(continued)

216

Central Pain Syndrome

TABLE 6.3 (continued)

Author(s)

Type of pain

Results/notes

Lazorthes et al. (1995)

Includes all pts operated on and
previously published by both
Lazorthes and Siegfried

SCI pain (101 pts)

Barolat et al. (1995, 1998)

SCI pain (11 pts)

Peyron et al. (1998)

Anderson and Burchiel (1999)

CPSP (Wallenberg) (3 pts,

with evoked pain)
CPSP

SCI pain included traumatic

paraplegia pain, iatrogenous lesions
or following cord tumor surgery,
herpetic myelitis and spondylotic
damage. Successful pain relief:
Short-term: 5058% of pts;
long-term: 3034% of pts
Authors conclusions: cord CP and
even more BCP respond poorly to
SCS, with increasing degrees of
denervation. Analgesia is much less
significant for SCI CP or iatrogenic
CP following surgery on the cord
(e.g., for tumor). Failures due to
degeneration of lemniscal fibers
Short-term successful pain relief:
45% of pts. 55% of pts never
experienced any pain relief (half
never felt paresthesias in the painful
area) Long-term successful results
only in 27% of pts, with good
(450%) pain relief in 2/11 pts and
moderate (2550%) pain relief in
1/11 pts.
Authors conclusions: results of SCS
on SCI pain have been disappointing
in the vast majority of pts
Failure

Tseng (2000)

SCI pain (1 pt)

Relief at 19 mos

Eisenberg and Brecker (2002)

Cord CP (postspinal cord

tumor removal) (1 pt)

Pain relief for 9 mos.

Above. Lesion SCS

Sindou et al. (2003)

Cord CP (30 pts; MS: 9 pts;

trauma: 7 pts; spinal tumor: 5pts;
syrinx: 5 pts; spondylotic
myelopathy: 4 pts)

Long-term results (mean follow-up:

18.8 mos, range 11.219.2 mos):
pain relief 450% (and minimal
drug use): 12/30 pts (40%)
All pts had incomplete spinal cord
damage (CP pts with complete
spinal cord damage or midline pain
excluded). SCS: paddle. Previous
TENS course, but results not given.
No differentiation between end-zone
pain and diffuse CP. At least some
retained sensibility in the painful
areas and normal or near normal
somatosensory evoked potentials in
most responders

Authors conclusions: CPSP is not

particularly responsive to SCS

Neuromodulation

217

Author(s)

Type of pain

Results/notes

Quigley et al. (2003)

Spinal cord/root compression

(4 pts)
MS (4 pts)
Paraplegia pain (3 pts)
19892000

Relief 50% in 4 SC-root compression, 3 MS and 0 paraplegia pain

(doctors assessment), 2 of 3, 2 of 3
and 0 of 2 (patients assessment).
General anesthesia, laminotomy in
most patients, 480% receiving a
quadripolar plate. Almost 60%
inserted at T912. Then C14,
C57, T58. 62% radiofrequency,
38% IPG.TEST: 5 days, retrospective
study via questionnaire. No routine
antibiotics. Majority of ALL patients
used the SCS every day for about
12 h, 21% only during
exacerbations, 10% did not use it
anymore. Average time from implantation to data collection: 4.2 years.
64 revision operations out of 102
pts, due to electrode complications,
generator complications,
connecting lead fracture. Global
infection rate was 4.9% (2 of 5 pts
needed explantation). Globally (CP
plus all other pains), pts who had
used SCS for 5 years or more had
lower levels of substantial pain relief
compared to those using it for less
(65% vs. 81%). It is unclear if this is
due to tolerance, an initial placebo
response, hardware failure or some
other phenomenon.

Rogano et al. (2003)

CCP (12 partial lesion pts)

VAS from 9.9 to 3.6 (no details

given). Minimum follow-up: 6 mos
(mean 19.1 + 13.5 mos)

Kumar et al. (2006)

Includes all pts operated on and
previously published by this group

MS CP (19 pts)

Initial pain relief: 17/19 pts

Long-term success (50100%
relief): 15/17 pts
Initial pain relief: 7/15 pts
Long-term success (50100%
relief): 5/7 SCI pts
Mean follow-up whole series
(including CP): 97.6 mos
Limb pain considered to be due to
cord injury. Favorable response in
cord lesion pts with incomplete
paraplegia and with the majority of
pain felt below the lesion level.
No benefit with SCS in pts with
complete paraplegia complaining of
either pain at the level of injury or
diffuse pain below the injury level

Kim et al. (2006)

CCP (cavernoma) (1 pt)

SCI pain (15 pts)

Failure

218

Central Pain Syndrome

clinical pain is small (less than 30%), and patients submitted to SCS all reported that
they felt some sensations, when in fact the stimulator was not activated. Even today,
there is a lack of high-quality evidence, no double-blind randomized trial (admittedly
rather difficult to set up in this context) and serious flaws in blinding, recruitment
and assessment in nearly all studies (Cameron 2004; Carter 2004).
When pain is below the lesion, SCS can be effective only if the corresponding
dorsal column(s) retain sufficient functional value. If the territory below the lesion is
totally anesthetic, SCS will not work. As a matter of fact, if the dorsal columns are
totally interrupted, electrodes even if implanted above the lesion cannot stimulate
the degenerated contained lemniscal fibers. Imaging and measurement of SEPs may
be useful to check integrity of the dorsal columns. Instead, SCS appears to be effective
in some patients with incomplete lesions, painful spasms, at-level pain or
postcordotomy pain. Poor results are seen with complete lesions and intermittent
and burning pain. Most studies report a decline in efficacy of SCS over time.
Generally, the best results have been obtained with multipolar electrodes, with
laminotomy epidural placement (Carter 2004), when electrodes are localized above
the pain segments, if stimulation paresthesias and pain segments are superimposed
and when the pain is localized rather than diffuse.
In conclusion, despite occasional spectacular successes, SCS is not indicated for CP
of brain origin and only a minority of well-selected CCP patients may obtain relief in
the long term (years) (see also Warms et al. 2002).
4. Transcutaneous electrical nerve stimulation (TENS) (Table 6.4)
TENS was first introduced in the 1960s as a screening procedure for SCS. It is applied
at high frequency (80100 Hz) (also known as conventional TENS) aimed at
activation of myelinated cutaneous sensory fibers or low-frequency stimulation
(short trains of impulses at 14 Hz over the motor nerves, known as acupuncture-like
TENS), aiming at activation of muscle efferents/cells and thereby evoking muscle
afferent input to CNS. Stimulation must be directed over the most painful region,
with dual-channel stimulators to cover a large body area with pain.
Mechanism of action. TENS can apparently reduce CP only if the dorsal column
medial lemniscal pathways are uninjured or only mildly injured (i.e., paresthesias are
evoked), perhaps by segmental conduction block of spinal projection fibers. At
appropriately high stimulation frequencies, after-hyperpolarizations seen in dorsal
horn neurons could coalesce and maintain the cell in a hyperpolarized and, therefore,
inhibited state (just as PNS would on neuromas) by tetanic hyperpolarization.
However, in the clinical situation, the intensity of the stimulating current for pain
relief is commonly below the threshold for activation of C- and A-d fibers, and the
relief may last days and occasionally weeks. Also, if central sensitization renders lowthreshold afferent input painful, it would be hard to explain how augmentation of
such input through TENS (or SCS) would suppress the pain, and, in fact, TENS may
exacerbate CP during stimulation. Suprasegmental mechanisms are, however,
possible (Sjolund 1993).

Neuromodulation

219

TABLE 6.4. Transcutaneous electrical nerve stimulation (TENS)

Author(s)

Type of pain

Results/notes

Banerjee et al. (1974)

SCI pain (5 pts)

100% relief at short term (30 min

three times per day)

Long and Hagfors (1975)

Pain secondary to CNS injury

TENS relatively ineffective

Davis and Lentini (1975)

SCI CP (11 pts) plus other

SCI neuropathic pains

2 successes, 2 partial successes,

18 failures; 4/4 failures for cervical
lesions, 5 successes and 6 failures
for thoracic lesions and 50%
success for conus-cauda lesions

Hachen et al. (1978)

SCI pain (39 pts)

49% early success, 28% late

(3 mos) successes

Heilporn et al. (1978)

SCI pain (3 pts)

Failures

Guilmart (thesis, detailed in

Sedan and Lazorthes 1978)

Brain CP (2 pts)
SCI CP (9 pts)

1 relief
Failures
(Conventional TENS)

Long et al. (1979)

CP of any origin

Patients with CP of any origin do not

respond to TENS in significant
numbers and responders do not
seem to maintain the response over
a long period of time. TENS usually
worsen hyperesthesia

Eriksson et al. (1979, 1984)

BCP (7 pts), CCP (11 pts)

In 6 pts acupuncture-like TENS,

conventional in others.
BCP: pain relief (continued for
3 mos.) in 5
CCP: pain relief at 3 mos in 7 (in 6,
at-level pain, not below-level CP)
Successful pain relief probably in
incomplete lesions
Not broken down from group:
probably some reliefs

Brainstem facial CP (5 pts)

Sindou and Keravel (1980)

Thalamic pain (5 pts)

Cord CP (17 pts)

Failures
Relief in 2 (late follow-up not
specified)

Bates and Nathan (1980)

Thalamic CP (12 pts)

8 stimulated beyond 1 wk.

Stimulation up to 8 h daily;
frequency up to 70 Hz. 0/8 helped
by TENS. Pts did not notice any
interaction between the sensation
and their pain, except that when the
intensity of stimulation was
increased, this suddenly added to
their pain.
(continued)

220

Central Pain Syndrome

TABLE 6.4 (continued)

Author(s)

Type of pain

Results/notes

Cord CP (16 pts; postcordotomy:

2 pts; intrinsic spinal cord
lesions: 8 pts; syringomyelia and
syringobulbia: 6 pts)

10 stimulated beyond 1 wk.

Detailed results not given

Globally, of 235 patients with

chronic pain and 160 passing test,
2025% used TENS at 2 yrs or
more of follow-up, sometimes only
to help them over crises of pain
Leijon and Boivie (1989b)

CPSP (infratentorial lesions)

(15 pts)

Pain relief from conventional or

acupuncture-like TENS in 4 (3 after
2 yrs). 3 pts (2 brainstem infarction,
1 unknown lesion site) continued to
report pain relief after 2 yrs. All 3
had normal or near normal
touch-vibration thresholds. One pt
with Wallenberg syndrome had
facial pain on one side and
extremity pain on the other.
High-frequency TENS for facial pain
used without effect on arm and leg
pains. High- and low-frequency
TENS had approximately equal
effect in the other 2 pts
The study applied rigid schedules
not taking into account the varying
distribution of pain and the
subsequent need to apply the
electrodes over the region
with the most intense pain.

Tasker (2001)

CP

TENS is seldom useful in pts with

pain over a wide area of the body. It
may be useful for pain in the
trigeminal area

Kabirov et al. (2002)

CCP (syrinx) (14 pts)

30100% relief in 12 (TENS

10 sessions, 60 min each)

Nuti et al. (2005)

CP: 410 pts

(including 3 Wallenbergs CPs)

No significant analgesia

Efficacy. Controlled trials are lacking, and so are long-term studies. While certainly
less expensive than SCS, DBS or MCS, and with almost no adverse effects, TENS
cannot cover wide body areas and requires prolonged use several times a day,
basically hampering a patients daily activities. While a trial may be warranted before
other more invasive procedures are contemplated, few patients gain long-lasting pain
relief, both with BCP and below-level CCP. TENS may relieve some SCI patients with
muscular or at-level pain. For MS spontaneous CP, TENS is ineffective (Rosen and

Neuromodulation

221

TABLE 6.5. Gasserian ganglion stimulation

Author(s)

Type of pain

Results/notes

Taub et al. (1997)

Taskers group

CPSP (brain 3 pts; brainstem 3 pts;

bulbar tractotomy: 1 pt)

Successful pain relief: 5/7 pts (100%: 1 pt; 75%:

1 pt; 5074%: 2 pts; 50%: 1 pt). The 2 failures
had an initial success which was lost within a
month (placebo effect?). Among these 5 pts, a
patient with thalamic infarct experienced relief for
21 mos and then found the stimulation was no
longer effective. Another stroke case found that he
no longer needed the stimulator because the pain
had subsided. Median follow-up: 21 mos. CP
better relieved than PNP in this unique series

Barsoum 1979; Young and Goodman 1979; Tasker 2001). Even cutaneous field
stimulation (16 metal pin skin cathode with single 1 ms, 4 Hz pulses, 30 minutes bid
at twice the sensory threshold) in TENS-resistant (low and high frequency) adds little
(20% more relieved patients) in PNP and maybe CP at 3 months.
The recently introduced scrambler therapy has yet to be tried on CP (Marineo
2003).
5. Gasserian ganglion stimulation (Table 6.5)
This was introduced in 1978 by Steude (see in Meyerson and Linderoth 2001).
Presumably, the efficacy depends on an intact afferent pathway in the periphery along
which nerve impulses generated by stimulation can reach the trigeminal nuclei in the
brainstem and continue transsynaptically up to the cortex. Its place in the treatment
of CP is virtually nonexistent.
6. Vagal nerve stimulation
There are no reports as far as CP is concerned. Given its possible unspecific effects on
catecholamines (Kirchner et al. 2001), this technique is likely not to have an impact
in the treatment of CP.
7. Electroconvulsive therapy (Table 6.6)
Introduced in 1938 by Ugo Cerletti, this has also been employed for pain control (see
complete bibliography in Canavero and Bonicalzi 2001).
Mechanism of action. Bilateral ECT sends electric impulses through the thalamus,
the hypothalamus and the brainstem. While ECT affects many neurotransmitters
and neuroendocrine substances (e.g., endorphins, acetylcholine, (nor)epinephrine,
dopamine, serotonin and GABA), Salmon and colleagues (1988) found no significant
correlations between endorphin levels and ECT in CP; they also noted no placebo
effect. Such changes of neurotransmitters but also changes in gene expression
repeated over the course of a series of ECT treatments could modulate neural
function at numerous sites throughout the nervous system. The a4 subunit of GABA

222

Central Pain Syndrome

TABLE 6.6. Electroconvulsive therapy (ECT)

Author(s)

Type of pain

Results/notes

Von Hagen (1957)

CPSP (thalamic) with evoked pains

and depression (1 pt)

Great improvement for about 10 mos from 8

bilateral ECTs, then relapse (1955). Further
pain control from 3 additional treatments.
Previous ECT for depression

White and Sweet (1969)

CCP (postcordotomy) (? pts)

Relief only during the confusional state

Salmon et al. (1988)

CPSP (thalamic) (4 pts)

Failure with unilateral ECT. No depressed pts

McCance et al. (1996)

CPSP (3 pts)

2/3 CPSP of immediate onset. 3/3 pts with

allodynia. 1/3 depressed pt.
(Few months CP remission in 1 pt after an
epileptic fit)
A course of six bilateral ECT sessions over
2 weeks slightly improved CP only in 1 pt, while
2 worsened

Doi et al. (1999)

Brain CP (12 pts)

Abstract. CP remission in 1 depressed pt after ECT.

Bilateral ECT (110 V for 5 min) for 612 sessions
at 17 day intervals. Complete relief of both
steady and evoked pain in all suprathalamic cases.
Partial relief in thalamic cases. Pain recurrence
relieved by a new ECT course in 9 pts

Harano et al. (1999)

CPSP (thalamic pain) (39 pts)

Abstract. Convulsions (plus nausea and

vomiting) lasting 23 min induced by
intracisternal (cerebellar) methylprednisolone
sodium succinate 125 mg in 5 ml syringe
mixed with CSF. Excellent results in 54.4%,
good in 38.6% and poor in 4%.
Lateral position; 22G 6 cm block needle
inserted at crossing point of bilateral mastoid
line and sagittal halfline under fluoroscopy.
57 injections in 39 pts

Fukui et al. (2002)

CPSP

Relief with bilat. ECT (paper not available for

review)

Canavero and Bonicalzi

(2003)

Cord CP (1 pt)

No pain relief after injection of 125 mg of

methylprednisolone in the lateral ventricle. No
frank fit

A receptors may be implicated in the clinical effects of ECT (see in Olsen and Avoli
1997).
ECT likely has direct, acute effects on the cerebral cortex. In the words of Von
Hagen (1957): Electroshock therapy may produce its effect . . . from a reduction in
the influence of the cortex on . . . reverberating . . . (circuits). We proposed that ECT
interferes with a corticothalamic reverberation mechanism (Canavero 1994;
Canavero and Bonicalzi 2001). Seizures may be a natural example of spontaneous
ECT: case 3 of Bornstein (1949) reported that a phantom sensation slowly shrunk
before an epileptic fit to recede totally at the moment of the fit. After recovering

Neuromodulation

consciousness, the phantom reappeared only after a certain lapse of time, a possible
sign of the warm-up period required by the reverberation to restart.
The minimal electrical intensity needed for a generalized seizure of a specified
minimal duration appears to vary by approximately 40-fold in the population
(Sackeim et al. 1993): this range may also apply to reverberation strength.
Efficacy. Some patients with CP have been meaningfully relieved by ECT for more
than a short time. Given the high rate of relapse (perhaps particularly in previously
drug-refractory cases), the need for multiple courses, possible permanent side effects
(amnesia) and non-uniformity of response, ECT should be considered as a last resort
in highly refractory cases.

8. Conclusions
The most important paper providing conclusive evidence about the role of electrical
neurostimulation for CP is that of Katayama and colleagues (2001). These authors
analyzed a series of 45 patients with CPSP, all tested with percutaneous SCS.
Satisfactory analgesia was set at 60% reduction on a VAS scale. In the long term
only 7% (3 patients) achieved satisfactory analgesia with SCS. Of the remaining 42,
12 underwent Vc DBS (in 7 also of IC and/or medial lemniscus): 25% (3 patients)
obtained satisfactory relief in the long term, while 31 patients in whom SCS was
ineffective underwent MCS (1 underwent both MCS and Vc DBS): 48% (15 patients)
obtained long-term relief. In particular, 9% (3/35) thalamic-infrathalamic and 0 of 10
suprathalamic obtained long-term relief from SCS; 0 of 2 suprathalamic and 30%
(3/10) thalamic-infrathalamic obtained long-term relief from DBS; 37.5% (3/8)
suprathalamic and 52% (12/23) thalamic-infrathalamic obtained long-term relief
from MCS. In sum, CS is superior to all other techniques.
Thus, in light of minimal invasiveness, no reported mortality and disabling
permanent morbidity, and the possibility of running placebo tests, CS is the
technique of first choice in BCP patients in whom oral drugs as previously suggested
have proved ineffective. In CS failures, DBS with simultaneous implantation of Vc
(or ML) and PVG should be attempted as a second step. For CCP cases with some
retained sensibility in painful areas, SCS is the first choice; in failures or totally
anesthetic patients, there is as yet not enough evidence to support CS over DBS. CS
may be a first option for reasons discussed above. TENS may be an option if cost is at
issue or patients refuse surgery.
These conclusions must be tempered by the expensive nature of such treatments,
including changes of batteries, loss of efficacy in several to many patients depending on the technique and explantation for intercurrent problems. Whereas SCS
and CS appear to be safe, DBS carries a small risk of mortality and disabling
morbidity.

CHEMICAL (Table 6.7)

Spinal administration of several drugs has been spearheaded by the not-wellunderstood observation that drugs ineffective by the systemic route often are effective

223

224
Type of pain
SCI pain

SCI pain
CP, thalamic (2 pts)

CPSP (1 pt)
CCP (15 pts)

CPSP (3 pts)

Author(s)

Pollock et al. (1951)

Davis et al. (1954)

Waltz and Ehni (1966)

Namba et al. (1984)

Glynn et al. (1986)

Crisologo et al. (1991)

In all, complete or almost complete

sensory block
Case 1: thalamic stroke with left hemisoma CP;
6 mos later, left stroke with right hemisoma
pain. IT lidocaine: at 0.5%: 0% relief; at 2%:
100% relief in left leg for 5 h
Case 2: right hemispheric cortical stroke with
CP in left arm/leg. Lidocaine at 0.5%: 0%
relief; at 2%: 100% relief for 1 h, then gradual
relapse at 5 h

Non-RCT, single-blind, crossover, single-dose

study. EPI clonidine vs. EPI morphine. Pain
relief: EPI clonidine: 7 pts (morphine
unresponsive); EPI morphine: 5 pts
(3 clonidine responsive). 3 pts unresponsive
both to morphine and clonidine, 2 of them
buprenorphine responsive

EPI clonidine (150 mg)

EPI morphine (5 mg)
EPI buprenorphine (0.3 mg)

IT lidocaine (0.5%, 2%, 2 mL)

Failure

Immediate abolition of leg pain, even before

sensory block. In one case, leg pain was
abolished, while arm and face pains were
reduced

Completely relieved spontaneous, diffuse,

burning, below-level pain

In a number (unspecified) of cases, spinal

anesthesia below level: burning pain did not
disappear. In 4 cases with CSF block,
anesthesia above level: in 3 distal pain
gradually disappeared, then slowly returned
(in 1 case, absent for 2456 min, full relapse
at 3 h)

Results/notes

IT morphine

IT pantocaine (6 mg)

IT local anesthetic

IT tetracaine 1 ml (0.5%)

Drug

TABLE 6.7. Chemical neuromodulation (spinal intrathecal [IT] or epidural [EPI] infusion)

225

CCP (MS: 4 pts; spinal cord compression: 1 pt; transverse myelitis: 2 pts)

Herman et al. (1992)

SCI (2 pts)

SCI (21 pts)

Loubser and Donovan (1991)

IT baclofen (50 mg)

IT lidocaine; 50100 mg (2 injections 1

h apart)

(continued)

RCT. Spontaneous burning pain and intermittent sharp pain. IT lidocaine effects: (1)
sensory
level of anesthesia above the level of injury in
pts with lumbar and thoracic injuries and to T4
in pts with cervical injuries; (2) significant
reduction of pain intensity when compared with
placebo (13 vs. 4); (3) analgesia lasting for a
mean time of 123 min, exceeding the expected
duration of action for interruption of nociceptive messages. IT lidocaine effects on pain:
overall: 65% relief of pain (mean) in 12/16 pts
Pts with spinal canal obstruction, sensory block
above SCI level: no change in 4 and 20% relief
of pain in 1. Negative response in 4 pts (2 with
incomplete anterior cord syndromes), despite
sensory anesthesia rostral to the level of SCI
(pain generator more rostrally?)
When spinal anesthesia proximal to SCI level
was adequate, 9 of 11 had a positive response
vs. 4 of 10 who did not obtain anesthesia
above SCI level, because of spinal canal
obstruction or high lesion level
Cord CP: RCT (crossover with placebo
vehicle) assessing the efficacy of acute IT
baclofen on chronic, dysesthetic and
spasm-related pain. IT baclofen significantly
suppressed dysesthetic pain and, after
the suppression of neuropathic pain,
spasm-related pain
Non-RCT

Case 3: thalamic CPSP (longer duration and

higher intensity than cases 1 and 2): 0% relief
at both concentrations, despite complete
sensory block

226
Type of pain
CCP (6 pts)

CCP, ASAS (1 pt)

SCI CP (1 pt)

BCP (thalamic CP: 3 pts; CNS injury:

1 pt)
CCP (paraplegia pain: 1 pt;
postcordotomy pain: 1 pt)
SCI pain (12 pts)

CPSP (8 pts)
SCI pain (6 pts)

Author(s)

Glynn et al. (1992)

Triggs and Beric (1992)

Loubser and Clearman (1993)

Reig (1993)

Fenollosa et al. (1993)

Taira et al. (1994, 1995)

TABLE 6.7 (continued)

Non-RCT. Pain and spasticity improvement

(450% relief): 8/12 pts. Minimal tolerance in
6/8 pts (after 3 yrs final dose range:
1.66.0 mg/day)
Substantial pain relief starting 12 h after a
single injection and persisting for 1024 h in
9/14 pts (3 SCI). Allodynia and hyperalgesia, if
present, also relieved. Placebo when tried
ineffective. Incomplete data on CP
components.
Study prompted by a CPSP suppressing effect
from 25 mg of IT baclofen in 1 pt with spasticity
(not relieved by baclofen) and pain

IT baclofen (50100 mg)

At 3 yrs of follow-up: never 475% relief. None

returned to work. 5075% pain relief: some;
unsatisfactory pain relief: some (numbers not
clear).
Congress abstract

IT morphine (initial dose, 1 mg, final

dose 3.4 mg/day)

IT morphine (0.31 mg/day,

continuous infusion)

Dysesthetic and cramping pain in both arms

and legs following a C6 incomplete injury. IT
lidocaine produced a sensory block to light
touch to the T8 level, with disappearance of
both spasticity and pain

IT lidocaine (50 mg)

Failure

Pain relief 50%: 3 (all with spasm). IT

clonidine: excellent pain relief in 1 pt. Better
relief with higher clonidine concentrations in
the CSF

EPI clonidine (150 mg)

IT clonidine (1 pt)

IT morphine

Results/notes

Drug

227

SCI CP (1 pt)

SCI pain (12 pts: 7 at-level pain and 2

below-level CP; musculoskeletal in
6 pts also present)

Cord CP (1 pt)

Thalamic pain (1 pt)

Paraplegia pain (6 pts)

Hassenbusch et al. (1995)

Loubser and Akman (1996)

Middleton et al. (1996)

Winkelmuller and Winkelmuller (1996)

IT opioids (implanted pump)

(continued)

Mean follow-up: 3.4 yrs (range 6 mos to

5.7 yrs).
Continuing IT opioids infusion 6 mos after the
pump implantation: 1/1 thalamic pain and
3/6 paraplegia pain pts. Initial mean morphine
dosage: 2.6 mg/day; at the first follow-up:
3.6 mg/day; at the last follow-up: 5.2 mg/day
No separate analysis results of BCP/CCP.

Non-RCT. Effects on neurogenic pain at both

6- and 12-mos interval: no significant change
in pain severity in 7/9 patients; pain increase
in 2/9 pts. Significant decrease of
musculoskeletal pain (5/6 pts).
Authors conclusions: IT baclofen does not
decrease SCI CP. Results of other studies were
possibly positive due to higher doses achieved
by bolus injections and continuous infusion
resulting in comparably lower CSF doses;
moreover, pain relief was assessed over only
24 h
Anterior cord syndrome case with incomplete
C5 tetraplegia.
Symptoms not improved by the administration
of IT baclofen through an existing
programmable infusion pump. Immediate pain
relief after clonidine was added to baclofen in
the pump reservoir and combined IT
administration started

IT baclofen infusion (implanted pump)

IT baclofen and IT clonidine

NRS reduction from 9/10 to 5/10 1 mo after

the pump implant. At 2 yrs follow-up, NRS
6/10 in spite of IT sufentanil trial and oral
propoxyphene addition. At last follow-up, pain
relief judged fair (25%) by the pt and a failure
by the authors. Positive preimplantation test

IT morphine (0.2 mg/h)

(IT sufentanil)

228
Type of pain
SCI CP (8 pts)

Cord CP (syrinx) (1 pt)

Cord CP (ischemic myelopathies: 5 pts;

MS: 2 pts; post-traumatic myelopathies:
3 pts)

CPSP (1 pt)
CCP (2 pts, 1 syrinx)
CP (CPSP?) (2 pts)

MS incomplete T5 CP (1 pt)

Author(s)

Meglio (1998)

Angel et al. (1998)

Nitescu et al. (1998)

Anderson and Burchiel (1999)

Belfrage et al. (1999)

Becker et al. (2000)

TABLE 6.7 (continued)

Reduction of spontaneous and evoked pain.

Results not broken down according to pain type
(CP vs. other pains)
Complete pain relief for 20 mos. Pain
reappearance soon after baclofen
discontinuation (pump explanted on patients
request after progression of MS)

IT adenosine

IT baclofen (110 mg/day, continuous

administration) (450 mg at each refill)

Outcome of CP patients (out of 30 sundry pts)

not specified, but all 3 had 450% relief at
test injection

Non-RCT. Drug dosage: morphine 0.5 mg/ml,

buprenorphine 0.015 mg/ml, bupivacaine
4.755.0 mg/ml. Daily volumes tailored to
give the pts satisfactory to excellent
(60100%) pain relief, with acceptable side
effects
Results: MS-related pain: effective; ischemic
and post-traumatic myelopathy: ineffective in
5/8 (63%) pts (due to centralization at
higher levels of pain). Several refused to
continue treatment

IT opioids (morphine or buprenorphine)

and IT bupivacaine

IT morphine

Initial IT morphine dosage: 0.5 mg/day; 2 yrs

later, the pt needed 3 mg/day of IT morphine to
maintain the best possible analgesia (VAS
reduction from 10 to 2)

Baclofen failure.
Relief in 3, then 2 (due to side effects in 1)
with 450% relief at 1 yr
Average morphine dosage: 3 mg/day
At- and below-level pains not distinguished

2 pts: IT baclofen (50 mg)

5 pts: IT morphine
1 pt: both
Test: 0.5 mg IT morphine
IT morphine

Results/notes

Drug

229

IT morphine (0.75 mg (mean) (IT, bolus

0.21.5 mg) and/or IT clonidine
(50 mg (mean) (IT, bolus 50100 mg or
300500 mg over 6 h). Combination:
half of each dose. Minimum
4 injections, 1 day apart

SCI CP (15 pts) (below level: 13;

at-level: 4 pts; both types: 3 pts.
Figures not in agreement!)

SCI CP (1 pt)

SCI CP (2 pts)

MS CP (1 pt) plus other 2 chronic

pain pts

Siddall et al. (2000)

Siddall et al. (1994)

Ridgeway et al. (2000)

Penn and Paice (2000)

Ineffective. Very serious side effects. Infusion

stopped. Coma. Residual memory impairment

IT ziconotide up to 5.3 mg/h

(continued)

No relief at end of trial. 47% CP decrease at

14.4 mg/day. No further decrease at 28.8 mg/
day. Dramatic pain increase over time,
requiring an increase in concurrent opioid
administration. Trial stopped and IT baclofen
restarted after appearance of confusion and
sedation

6-day double-blind crossover, placebocontrolled RCT. Overall pain relief (4 h after

drug administration):
IT morphine alone IT clonidine placebo.
IT morphine (median minimal effective dose
0.75 mg) IT clonidine (median dose 50 mg
as bolus injection or 300500 mg over 6 h)
produced significantly more pain relief than
placebo 4 h after administration.
Pain relief 50% (mixture): at-level pain: 50%
of pts; below-level pain: 35% of pts (in this
group of patients, IT placebo was pain relieving
in about 30% of cases).
Conclusions: at-level pain appears to be more
responsive. The concentration of morphine in
the cervical CSF and the degree of pain relief
correlated significantly, so drugs should be
administered above-level
NNT: 7.5 (combination)
Pain unresponsive to IT morphine alone.
Marked decrease in pain from IT morphine IT
clonidine combined administration

Failure

IT ziconotide (and opioids

coadministration) up to 144 mg/day

IT morphine (10 mg/day)

& IT clonidine (17 mcg/day)

IT morphine (up to 3 mg/die)

Gatscher et al. (2002)

230
Type of pain
SCI pain (low lesions) (14 pts) (plus 19
cauda lesions pts)

BCP and CCP

CCP (18 pts), most spinal traumas

CCP (1 pt)

Author(s)

Margot-Duclot et al. (2002)

Canavero and Bonicalzi (1998, 2003)

Rogano et al. (2003)

Plus congress abstracts

Nuti et al. (2005)

TABLE 6.7 (continued)

IT morphine

IT morphine (16 mg)

Failure

VAS from 9.2 to 3.6, both in complete and

incomplete lesions. Minimum follow-up: 6 mos
(mean 19.1 + 13.5 mos).
No details are given and data are inserted
shortly in discussion. No differentiation
between at- and below-level pains. Follow-up
short. Nausea and vomiting frequent

Analgesia from IT midazolam correlates with

positive propofol test. Pump implanted in a few
pts. Satisfactory analgesia, although tolerance
may occur. Follow-up is entering a few years.
No side effects observed to date

Placebo-controlled study. 8 had 460% relief;

5 implanted with pump. Effect lost in a few.
Cauda pts: 12 had 460% relief and 10
implanted. Paroxysmal component more
responsive than steady pain

IT baclofen (50150 mg)

IT midazolam (2.56 mg/day)

Results/notes

Drug

Neuromodulation

when given spinally (references in Siddall 2002). Unfortunately, although several

drugs have been administered intrathecally (IT) in attempting to treat CP, there is
only a small number of papers reporting the effect of continuous IT administration
of drugs on CP and the vast majority of them deal with CP after SCI. These studies
are not randomized, nor controlled, and often patients with CP are no more than 1
or 2 cases among several other pain conditions or just single case reports. In most
papers, only the outcome of the mixed group of pain patients is reported and the
outcome of patients with CP remains unknown. Well-designed studies with
homogeneous groups of patients with long-term follow-up are needed before
drawing definite conclusions on any of the reviewed drugs. Moreover, a positive
preimplantation test does not guarantee long-term relief.
A review of the literature and of personal experience suggests the following
conclusions:
1)

2)

3)

4)

5)

6)

IT lidocaine significantly reduces pain in a proportion of SCI patients, if

access to the cord cephalad to injury level is preserved; however, relief may
not be obtained despite a sensory block above the level of injury. Although
good relief can be obtained, the effect is only temporary and even multiple
local anesthetic blocks do not result in long-term relief of SCI pain.
IT midazolam (a GABA A agonist) has significantly relieved several patients
with both BCP and CCP in our experience, without side effects of any kind,
although tolerance can be seen.
IT baclofen relieved few patients of their CP in the long run, as relief is often
lost (tolerance) due to long-term receptor changes (down-regulation at, for
example, cord laminas III or higher levels) or other factors. It may even make
pain worse in some patients. Although generally well tolerated, the general
impression is that it has no major effects on CP (see also Slonimski et al. 2004).
Clonidine (epidural or IT, but only poorly PO) is efficacious in some patients
with both BCP and CCP, but this effect may be no greater than placebo. Its
noradrenergic effects (a2 agonist) may modulate pain centrally; however,
there is no firm evidence that the analgesia due to activation of spinal
adrenoceptors is long lasting. In humans, long-term IT clonidine infusion
rarely produces pain relief beyond 3 months (Ackermann et al. 2003).
Epidural or IT morphine at a dose of 0.51 mg/day is initially effective against
SCI CP in some patients (particularly with incomplete injuries): at-level, but
much less below-level, pain appears to be responsive. The general impression
is that opioid efficacy in pure CP is poor, with rare patients drawing longterm benefit (similar to what is observed with oral drugs).
The IT calcium channel blocker ziconotide had proved of little value
(Bonicalzi and Canavero 2004).

Analgesia with all these drugs is due to targeting of spinal above-level or

supraspinal sites. Drug combinations may be more effective. Tolerance to a
combination of morphine and clonidine develops more slowly than with morphine
alone, but side effects are not reduced even with reduced doses of clonidine
(hypotension, sedation). While intermittent bolus and continuous infusion may not
differ in efficacy, infusion with a totally implanted pump is preferred to lower the

231

232

Central Pain Syndrome

infection rate, even if initially more expensive. The pharmacodynamics of IT-injected

drugs differs considerably on type of administration: a bolus dose produces much
higher concentrations of CSF baclofen compared to continuous infusion, particularly
at cervical and higher levels and a positive response to a bolus may not be duplicated
during continuous infusion. Also, spasticity and analgesia may require different
receptor subsets (Herman et al. 1992). An important caveat is that an excess of free
GABA may cause postsynaptic receptor changes, leading over time to desensitization.
We recommend that, if spinal infusion is selected, a combination of IT
midazolam/clonidine (on the basis of possible greater efficacy of GABA A modulation deduced by oral studies) or (in failed cases, due to GABA A downregulation)
baclofen/clonidine be the first option, particularly in patients with extensive
(hemisoma) CP, in whom CS may not ensure complete coverage of painful areas.
However, cases are on record where CS controlled the pain beyond expected
somatotopic limits.

PREVENTION
There are no known ways to preempt the development of CP, nor are there markers
for identifying pain-prone patients. Prophylactic amitriptyline does not appear to
exert any meaningful effect (Chapter 5). Neuroprotective agents have yet to deliver
the promise they raised in the stroke-trauma setting (Canavero et al. 2003).
Zimmermann (1979) questioned whether implantation of an electrical stimulator
immediately after a CNS lesion could prevent changes, like supersensitivity, leading
to CP. While this is of course totally unfeasible, brain reorganization could be
modulated by TMS and this may become a possibility in the future, once we identify
CP susceptibility markers. Preemptive manipulations, such as those employed in
amputees, e.g., with memantine (Flor 2003), to block SI reorganization and reduce
phantom pain, may also be explored for CP.

ALTERNATIVE APPROACHES
No systematic data are available for CP on any one technique.
Peripheral/regional and epidural neurolytic blocks (phenol, alcohol, anesthetics) are
basically useless in the long-term management of CP and some may be harmful:
results are short-lived or disappointing. However, abolition of oncoming normal
afferent stimuli can sometimes secure temporary relief; repeated or prolonged blocks
can dampen at least temporarily a patients suffering, sometimes for longer periods of
time than the duration of the block (Tasker et al. 1991). Since permanent surgical
neural interruption at the site of successful block usually fails to relieve the pain,
anesthetics likely act as pain modulators (Condouris 1976).
Psychologic support may be useful in selected patients as a corollary measure. Pain is
a highly intrusive event that is extremely effective at capturing attention. Cognitive
factors can alter the perceived intensity of pain and, accordingly, can modulate SI
activity in functional imaging studies (references in Schnitzler and Ploner 2000).

Neuromodulation

Cognitive strategies to deal with the situation may be used sometimes with
hypnosis as coping may change not just the perception of pain, but also autonomic
responses during noxious stimulation (Thompson 1981; Weisenberg et al. 1996). In
this regard, the relative increase of activity in the lateral orbitofrontal cortex during
pain may represent a source of cognitive modulation of emotional components that
are produced by or interact with pain processing (references in Petrovic and Ingvar
2002). Coping aims at making pain comprehensible, planning activities, taking drugs,
communicating and distracting oneself. One possible cognitive coping mechanism
involves suppression of activity in ACC, OFC and PAG: cingulate cortex, for one, is
known to regulate brainstem opioid network during opioid and placebo analgesia
(Petrovic and Ingvar 2002). However, in our extensive experience and that of others
(e.g., Warms et al. 2002), such strategies never provided substantial relief to CP
patients: submitting anyone to psychotherapy is callous and unrewarding, except to
help control depression, which may profoundly affect the perception of pain.
CP is life-long and a durable rapport with doctors is vital, particularly to rein in
moments of despair: thus, a placebo approach is warranted in all cases. For
instance, excellent interpersonal relationships, demonstration of caring by the
therapist and enthusiasm, spending time with the patient, supplying accurate,
rational information on the effects/results to be obtained, a predicted positive course,
belief in treatment efficacy and charisma (the surgical look) all affect placebo
circuits. Patients with strong dependency needs and desire to please will respond
positively, while those with more explicit conversion of negative affect and somatic
preoccupation respond negatively (Nicholson et al. 2002). All this is reduced by
informed consent, decreased physician paternalism/authority, and so on. When both
context and expectations are completely eliminated (hidden therapy), pain relief is
less than when therapy is in full view of the patient (Pollo et al. 2001). Anticipation of
pain relief is closely tied to the placebo response and intimately tied with actual pain
reduction. Since a high level of activity at prefrontal levels marks patients with high
expectation of pain relief and high levels of actual pain relief, prediction of response
to medication may become possible by looking at the expectation component in
patients brain scans. Also, the same sets of neurons activated both by experienced
and imagined (empathy) pain (ACC and other areas, but not SI) are also set in
motion by the anticipation of pain (Holden 2004 and references therein). However, it
should be stressed that in CP patients a placebo response seems less common
(unpublished observations) than other neuropathic pain states (Verdugo and Ochoa
1991), suggesting that placebo mechanisms may be disrupted in CP.
Riddoch (1938) already noted that CP could sometimes be diminished by
concomitant stimulation (e.g., pinching, induced itching, fractures); also, pushing
into the muscle tendons or bellies may relieve cramping pain up to a few hours
(McHenrys website: www.painonline.org).
Distraction from pain through attractive and pleasant hobbies is indicated, as these
compete for attention. We and others noted that orgasm can temporarily decrease
CP, but also vice versa. A distracting task can reduce pain by reducing activity in
somatosensory regions and the PAG. Cognitive distraction may attenuate the painevoked activity in the ACC, the insula and the thalamus, and in phantom pain this
can be activated by simple hypnotic suggestion (see review and references in Petrovic

233

234

Central Pain Syndrome

and Ingvar 2002). Pain can be momentarily soothed by making comparisons and
enduring the pain; consideration shown by others may also help.
Several alternative approaches have been attempted. However, trials of alternative
medicine must be considered in the light of their quality (as for any other therapy):
pot-pourris of several treatments such as EMG biofeedback, behavioral coping
training, cognitive behavioral therapy and the like may be moderately effective for a
short time (Edwards et al. 2000). On the other hand, several CP patients are poorly
compliant with their drug regimens (justifying some apparent failures) and it may
happen that, being under the doctors eye, the patient feels compelled to take drugs
on a regular basis and thus obtain drug-related benefit.
There is uncontrolled evidence that eastern medical treatments can allay CP,
particularly combined with western drugs (Yen and Chan 2003; Kong et al. 2004).
Li (2000) treated 20 cases of central pain after head injury (19901998) by
invigorating blood circulation. Bi Tong Tang (a decoction of several herbs taken
daily in divided doses for 14 days) was used for the pain. Acupuncture was used in
some patients for 7 days, and infrared radiation (20 minutes die) for 7 days. Pain
disappeared in 18 patients after 2 to 12 weeks of therapy, and in 2 was reduced.
Follow-up was not specified. Perhaps, this is the natural history of brain injuryrelated CP.
No reports exist on biofeedback techniques (surface EMG, temperature/thermal,
EEG-based) in the CP setting. A small controlled study found that people can learn to
suppress acute pain when shown the activity of the rostral ACC in real time from
fMRI represented on a computer screen as, for example, a flame of varying size in just
three 13-minute sessions, the effect seeming to last beyond the sessions in the scanner
(DeCharms et al. 2005). In this case, it would be important to define neurometabolic
markers of CP for possible image-guided feedback therapy.
Autogenic and/or progressive muscle relaxation training, physical and massage
therapy have only a minimal role to play in the vast majority of patients. However,
they may help in treating secondary or associated musculoskeletal and other
nociceptive components. Musculoskeletal pain arising from, for example, abnormal
posture must be specifically addressed in all cases. On the other hand, physical
activity may either increase or decrease CP in individual cases. Pain can be
momentarily soothed by changing body position.
Acupuncture has never relieved our patients (as also experienced by Bowsher
1994), a sign that acupuncture only works if the CNS is intact. However, it may
moderately help non-CP SCI pains (Warms et al. 2002).
Phantom pain has been reportedly reduced with mirror training: actually, only the
clenching spasm and cramping pain, not the burning or lancinating pains, were
reduced. Analogous treatment for CP has not been reported.
As regards CCP, it must be added that any factors that work to the detriment of
general health will often worsen or contribute to the severity of pain, and any form of
stimulation below injury level may worsen the pain (UTI, bladder stones, decubitus
ulceration, paronychia, stress, psychological factors, etc.).

Neuromodulation

A FUTURE?
Neuroablative techniques as a whole have failed to relieve more than a few patients,
often with intolerable side effects; sometimes, they have triggered new pains
(Chapters 1 and 7).
Today, properly used oral drugs can allay the suffering of many patients and totally
relieve a few. Neuromodulation as discussed may achieve similar results in the sizable
proportion of patients who drew no relief whatsoever, or help boost relief in others.
The bottom line is that only a minority of patients will not be helped by current
strategies (Table 6.8). On the other hand, patients with good initial relief may later
find that they can no longer tolerate even modest degrees of pain.
The literature is clear in this regard. For instance, in a SCI CP series (Falci et al.
2002), all patients were refractory to tricyclics, antiepileptics, baclofen, klonopin and

TABLE 6.8. Treatment of central pain: the TANG guidelines

A. ORAL DRUGS
Step 1: mexiletine (up to 1000 mg) gabapentin (up to 3600 mg); timeline: 1 month
Step 2: lamotrigine (up to 600800 mg); timeline: 3 months
Step 3: amitriptyline (up to 150 mg) (only brain central pain); timeline: maximum 3 months
B. NEUROMODULATION
BRAIN CP
Step 1: (only if propofol AND/OR TMS responsive)
Extradural cortical stimulation (12 paddles)
OR (if hemisoma or diffuse pain)
IT midazolam/clonidine or IT baclofen/clonidine
Step 2: Bifocal DBS (Vc and PVG)
Step 3: Convulsive therapy
CORD CP
A. Some preservation of lemniscal conduction
Step 1: SCS
Step 2: IT midazolam/clonidine or IT baclofen/clonidine
Step 3: Extradural cortical stimulation (only if propofol and/or TMS responsive)
OR Bifocal DBS (Vc and PVG)
Step 4: Convulsive therapy
B. No preservation of lemniscal conduction
Step 1: IT midazolam/clonidine or IT baclofen/clonidine
Step 2: Extradural cortical stimulation or Bifocal DBS (Vc and PVG)
Step 3: Convulsive therapy
* If elected, TENS should be attempted at this time combined with drugs.

235

236

Central Pain Syndrome

opioids (at most taking the edge off the pain), and in some to IT opioids, baclofen,
clonidine, local anesthetic or SCS.
The promise of peptides (e.g., neurotensin; cholecistokinin modulation during
opioid administration; growth factors antagonists) coming from animal studies has
yet to materialize.
What options are available? It will be our contention that CP can be abolished,
immediately and permanently, by a small focal lesion in the internal capsule
(Chapters 7 and 8). We will try to prove that this is the only ablative technique with a
place in the therapeutic armamentarium of CP particularly at a time when neural
reconstruction with stem cells or other engineered cell lines or implantation of cells
secreting analgesic substances subarachnoidally in the nervous system is slowly
becoming a reality (e.g., Kondziolka et al. 2002; Wirth et al. 2002; Szentirmai and
Carter 2004; Fouad and Pearson 2004; Bang et al. 2005).
However, for those unwilling to undergo demolitive surgery of any kind, the
benefit of combining different strategies (despite higher costs) may represent an
interesting avenue. Recently, SCS analgesia has been found to be boosted by
concurrent pump infusion of baclofen or adenosine in PNP patients over a few years
(Lind et al. 2004). In CP, combining CS with IT drugs as discussed above could prove
effective, and so could the combination of CS with DBS (although at a high cost).
Given the pivotal role of GABA, the infusion of a GABA A agonist such as muscimol
(a drug with a 1520 minute posteffect seen in tremor patients; Levy et al. 2001)
directly in the thalamus or cortex may become an option, and so may the
transplantation of GABA (stem) cells. Mark and Tsutsumi (1974) have already
reported on intrathalamic infusion of lidocaine in the treatment of chronic pain.

PATHOPHYSIOLOGY: HUMAN DATA

To wrest from nature the secrets which have perplexed philosophers in all ages, to track to
their sources the causes of disease, to correlate the vast stores of knowledge, that they may be
quickly available for the . . . cure of disease these are our ambitions. (William Osler)
A theory that accounts for all the facts is bound to be wrong, because some of the facts are
bound to be wrong. (Francis Crick)

A theory of CP must be able to explain:

1)
2)
3)
4)

5)
6)
7)
8)
9)

its idiosyncratic character, i.e., why it arises only in some individuals and
not in others with apparently identical lesions;
its immediate or delayed (even for years) onset;
why both a small lesion of the dorsal horn of the spinal cord or a huge infarct
of the parietal cortex can equally trigger CP;
its continuous, spontaneous nature in the vast majority of patients, but also
its evoked components (allodynia, hyperalgesia) which, in some instances,
can be the only or opening symptom, plus radiation and prolonged
aftersensations;
its many different qualities, even simultaneously (including dysesthesias and
pruritus);
referral to superficial and/or deep structures;
pain intensity fluctuations, from day to day or month to month, for no
obvious reason and increases by both somatic stimuli and emotion;
constant somatotopical referral of pain to areas of sensory loss;
differential response of patients with apparently identical lesions, i.e., of
the same size, site and nature, to some treatments but not others.

In the end, the winning theory is the one that leads to a cure. In this sense,
all past and present theories fail.
We will make no reference to animal studies. In spite of veterinary evidence
of classical thalamic pain following Vc lesions (e.g., Holland et al. 2000), a critical
review clearly shows that current animal models of CP are wholly unsatisfactory to
say the least (see reviews by Willis 2002 and Pioli et al. 2003): not unexpectedly,
no therapeutic breakthrough ever came from such studies.
237

238

Central Pain Syndrome

Neuroanatomy and neurochemistry differ in humans and experimental animals.

For instance, substance P has been considered a key substance in pain transmission
on the basis of animal data, but the majority of clinical trials with human
NK1 receptor antagonists for a variety of acute and chronic pain states (including
migraine) gave negative results. Heat hyperalgesia, so commonly seen in animal
models, is only present in a small proportion of patients suffering from CP.
Even monkeys differ from humans, for instance in cognitive processing of pain.
Autotomy, a classic sign of pain in animals, has been reported in several human
patients without pain (e.g., McGowan et al. 1997; Tasker 2001). In the words of
Gazzaniga (1998):
Humans often turn to the study of animals to understand themselves . . . monkeys . . . It
has been a common belief . . . that the brains of our closest relatives have an organization
and function largely similar, if not identical, to our own. Split-brain research has shown
that this assumption can be spurious. Although some structures and functions are
remarkably alike, differences abound. The anterior commissure provides one dramatic
example . . . When this commissure is left intact in otherwise split-brain monkeys,
the animals retain the ability to transfer visual information from one hemisphere to the
other. People, however, do not transfer visual information in any way. Hence, the same
structure carries out different functions in different species an illustration of the limits
of extrapolating from one species to another. Even extrapolating between people can
be dangerous.

Actually, animal data in the whole field of biomedicine show profound flaws (see
review by Pound et al. 2004; Linazasoro 2004). Even capsaicin-evoked pain in human
volunteers is not a model of neuropathic pain, as the latter is often delayed
and generally permanent, whereas capsaicin-induced hyperalgesia develops within
minutes and is transient.
Finally, CP differs from peripheral neuropathic pain (PNP) and, although they
share clinical similarities, no attempt will be made to correlate the two.
The vast majority of theories proposed to explain CP until now are based on
incomplete current anatomical knowledge, selective adaptation of anatomical
data (often of animal provenance) to authors needs, scarce appreciation of the
full clinical spectrum of CP and its features, exclusion of contradictory findings
or scarce familiarity with the full gamut of neurosurgical data (e.g., Melzack 1991;
Cesaro et al. 1991; Jeanmonod et al. 1996; Craig 1998). Some are technology- more
than idea-driven. Dismissal of exceptions not just single cases, but whole groups of
patients is the norm in the field. It should not come as a surprise that different
authors, based on similar evidence, reached opposite conclusions. As physicist
Stephen Hawking (1988) put it, you can disprove a theory by finding even a single
observation that disagrees with the predictions of the theory . . . if ever a new
observation is found to disagree, we have to abandon or modify the theory. At least
that is what is supposed to happen. Finally, a few authors embarked on phrenological approaches that try to paste the CP sensation to a unique brain center, with
scarce success.
Our thesis is straightforward: CP is the end result of a decorrelation of sensory
information processing along the sensory thalamocortical loop. The only permanent
cure, bar complete neural restoration, is a stereotactically guided lesion of the

Pathophysiology: Human Data

descending arm of this loop. This theory refutes neuromatrix approaches, which,
in light of the wide network of interconnecting areas at the basis of acute pain,
find surgical lesions useless (Melzack 1991). It also refutes the suggestion that the
spontaneous, resting component of CP is the end result of different pathophysiologic
mechanisms (see Garcin 1937): differential engagement of this single mechanism
explains, for instance, different descriptions of the pain. This chapter reviews the
evidence for such a theory.
Important sources of information regarding the genesis of CP are:
1)
2)
3)
4)
5)

functional imaging studies;

neurophysiologic studies;
results of neuroablation for CP;
reports of sudden disappearance of CP;
pharmacological dissection data (see Chapters 5 and 6).

These will be reviewed in detail.

Throughout the text, no reference to psychological theories of CP will be
made, not because of a dearth of such theories, but for the simple reason that CP
is somatic pain that cannot in any way be understood in terms of a psychological
(e.g., cognitive or psychodynamic) framework of any kind, but in reductionist
terms. In addition, several studies indicate that CPSP is not part of a psychiatric
disorder. A Danish group (Andersen et al. 1995) found no statistical evidence
of an association between depression, social factors or major life events and CPSP.
Mukherjii and colleagues (1999) found depression-dysthymia in 41% of CPSP versus
40% of non-CP stroke patients. Likewise, Stenager and associates (1991) found
no differences between MS patients with and without pain with respect to depression
(see also Osterberg et al. 1994). Thus, the presence of depression/dysthymia does
not correlate with CP. Even suicidal ideation, which is frequent (up to a quarter
of patients at some point of their history), is proportional to severity of pain
and hostility, and not depression. On the other hand, like all pains (and medical
conditions), the experience of CP may be influenced by so-called psychological
factors.

FUNCTIONAL IMAGING STUDIES

With the advent of human neuroimaging over the last 20 years, there has been
a trend to use this technique with its pretty pictures of colored blobs on brain slices
almost as a modern-day phrenology. It is crucial that we remember that these pretty
pictures can easily mislead us and that their interpretation needs to take into account
the wealth of scientific evidence obtained with different methods from humans.
There are many, sometimes quite small, populations of neurons with different
responses to different types of stimulus or event in brain regions which may not all be
revealed by functional neuroimaging, which rather reflects the average metabolic
demands of a brain region. Further, brain imaging does not address the issue of
the information that is represented by virtue of the different tuning of individual
neurons (which are the computing elements of the brain), and so does not provide

239

240

Central Pain Syndrome

the evidence on which computational models of brain function must be based.

It is thus very important to consider the results of human functional neuroimaging in
the light of what is known from complementary studies using, for example, neurophysiology and the effects of brain damage (Kringelbach and Rolls 2004).
Not all neurometabolic studies provide the same degree of information. Some
refer to the spontaneous component of CP; others assess the brain response
to allodynic conditions. A few address receptor anomalies. It is important to keep
these separate. For instance, the spontaneous, resting component of CP can only
be assessed with single-photon emission computed tomography (SPECT) or positron
emission tomography (PET) but not functional magnetic resonance imaging
(fMRI) as commonly used. Drugs with clear-cut pharmacodynamic profiles can
dissect neurochemical mechanisms by modulating the resting state and provide
crucial pathophysiologic information.
1. Studies assessing the spontaneous resting component and
its modulation
1. Laterre and colleagues (1988) studied with fluorodeoxyglucose (FDG)-PET in the
resting state (twice, with a 2 month interval) a woman who developed CPSP due
to a small right infarct at the level of the posterior putamen and posterior limb of
the internal capsule, with no visible extension into the thalamus on MRI. There
was right hypoperfusion (17% asymmetry), particularly at the level of the posterior
thalamic complex as well as in the putamen. No metabolic alterations were found
in the cerebral cortex.
2. Lee and colleagues (1989) studied six CPSP patients with technetium-99m
hexamethylpropyleneamineoxime (HMPAO) brain SPECT: 4 infarctions in the
thalamus and internal capsule and 2 hemorrhages in internal capsule-putamen (4 left,
2 right). Three patients showed thalamic lesions and these had decreased rCBF in
ipsilateral parietal (one bilaterally) and temporal cortex and one in frontal areas.
Extrathalamic lesions showed no cortical anomaly.
3. Tsubokawa and colleagues (1991) studied at 410 days after implantation
of a motor cortex stimulator 7 patients with CP with 131I-amphetamine SPECT. The
rCBF showed a marked increase (150200%) in the stimulated cortex and the
ipsilateral thalamic and brainstem area, along with pain abatement. The skin
temperature as assessed with thermography in the painful area increased to almost
the same level as that in the contralateral non-painful area.
4. Hirato and colleagues (1993) submitted to PET studies with 18FDG and a steadystate method with C15O215O2 nine CP patients. MRI and CT revealed definite
thalamic (3), putaminal (3), thalamoputaminal (1) and cortical parietal (2) damage.
Superficial pain was more marked in cases with definite thalamic damage. In patients
with a thalamic lesion, there were many irregular burst discharges in the Vop-Vim
area at stereotactic microrecording. The relative value of regional cerebral glucose
metabolism (rCMRGlu) decreased in the lesioned thalamus, but increased in
the cerebral cortex around the central sulcus on the lesioned side. However, the
relative value of regional cerebral oxygen metabolism (rCMRO2) did not increase

Pathophysiology: Human Data

(dissociated glucose/oxygen metabolism of the same area). In patients studied with

both techniques, OGMUR (oxygenglucose molar utilization ratios) in the premotor
area and SI/MI decreased more in cases with a thalamic lesion than in those
with a putaminal lesion. In a patient with a combined putaminothalamic
lesion, neural activity was reduced in the Vim-Vc area, with peripheral receptive
fields to electrical thalamic stimulation being predominantly in the face, hand and
sometimes in the foot area. In this case, the regional oxygen extraction rate (rOEF)
was markedly increased in the cerebral cortex around the central sulcus on the
side of the lesion, despite the chronic stage of cerebrovascular disease. In two
patients with cortical lesions, who showed mild superficial pain with or without deep
pain, rCMRGlu was decreased in the lesioned cerebral cortex. Though no ischemic
lesion could be demonstrated by CT, rCMRGlu was reduced in the lesioned Vc.
In patients with a subcortical lesion, rCMRGlu commonly decreased in this area.
Therefore, rCMRGlu in this area was decreased in all cases with CP including cortical
cases. This study then showed that OGMUR in the cerebral cortex around the central
sulcus was markedly decreased on the damaged side in cases with thalamic lesions.
However, in patients with a putaminal lesion, it was only moderately decreased,
particularly rostrally. In patients with subcortical lesions, the more severe
the superficial pain, the higher was the relative value of glucose metabolism
compared to that of oxygen (which was a reciprocal value of OGMUR) in the
cerebral cortex around the central sulcus on the involved side. In the patient with
combined lesions, rOEF was increased in the same area. Sensory thalamic hypoactivity (decreased rCMRGlu) was seen in all cases. In sum, in the thalamic lesion
group with pain (superficial pain dominant), r-O2 consumption was maintained in
most brain structures, except in the lesioned thalamus, while in the cortical central
sulcus this was normal, but the rO2 extraction ratio was increased and so was the
relative value of r-glucose utilization compared to r-O2 consumption. In the patients
with thalamic lesions and pain (deep pain dominant), both r-O2 consumption and
O2 extraction ratio were reduced in all brain structures and so was glucose
metabolism. They concluded that increased activity in SI/MI combined with a
decreased activity in Vc appeared to be a marker of CP, with character of pain
(superficial versus deep) depending on different processing at thalamocortical
levels (see also Svensson et al. 1997, who show that acute skin pain increases thalamic
CBF and decreases MI/SI CBF). The same group (Hirato et al. 1995) reported that
in one putaminal hemorrhage case (included in the above analysis) PET
renormalized after successful radiosurgical Vim thalamotomy.
5. De Salles and Bittar (1994) studied with FDG-PET a thalamic pain patient.
CP appeared two weeks following a stereotactic biopsy for a midbrain lesion and
worsened over one month. The patient complained of an annoying sensation of
needles and at times a burning sensation on the right hemiface and hand with
hyperesthesia to pinprick and light touch on the right face and hypoesthesia to
pinprick on the right fingertips. MRI disclosed that the needle had passed precisely
in Vc, plus the mesencephalon (where the medial lemnicus, which courses just
caudal to Vc, could have been damaged). Two months after the biopsy, PET showed
marked hypo/ametabolism of the left thalamic region, right cerebellar diaschisis

241

242

Central Pain Syndrome

and left parietal cortex hypometabolism. Ten

months later, allodynia with cold intolerance
persisted in the right hand and face. At this time
PET showed enduring thalamic hypometabolism,
recovery of the parietal cortex anomaly (which,
however, might be interpreted as a sign of
hyperactivity) and of the cerebellar cortex.
6. We (Canavero et al. 1995; Pagni and Canavero
1995; Canavero and Bonicalzi 1995; Canavero
et al. 1999) showed that patients with CPSP, CCP
(intramedullary cyst, syringomyelia) and other
CPs show basal parietal (SI) and/or frontal MI/
PM/PFC (in a few cases also temporal), plus
thalamic hypoperfusion on HMPAO and ECD
SPECT. These flow changes are promptly renormalized following successful treatment (propofol,
evacuation, cortical stimulation) (Figure 7.1).
7. The Lyon group (Peyron et al. 1995) reported
on 2 CP (both spontaneous and evoked) patients,
one with a right mesencephalic infarct with left
leg pain (spontaneous and evoked) and the other
with a left parietal infarct sparing the thalamus,
with right hemisoma pain, bar the face. In case 1,
PET at rest showed no cortical abnormality,
but right thalamic hypoperfusion (
9%).
Figure 7.1: High-resolution SPECT (double-head During MCS, CBF was increased in brainstem,
camera) images of postcordotomy CP. Note both orbitofrontal cortex (OFC), right thalamus and
thalamic (upper scan, arrowhead) and parietal cingulate cortex (CC): 30 minutes after discontihypoperfusion (lower scan, arrowhead).
nuation, persisting CBF changes were seen in
OFC and CC. In case 2, PET at rest showed
widespread CBF decrease in left parietal cortex (
35%) and hypoactivity in
left thalamus (
10%), this latter being normal on MRI. During MCS, CBF was
increased in brainstem, OFC, left thalamus and CC, while the parietal cortex
asymmetry was unmodified. Analgesic effects in both patients lasted at least
30 minutes after stopping MCS and this went along sustained CBF changes,
particularly in the thalamus. CBF increases were of the order of 79%.
An important sustained CBF increase was seen in patient 2s brainstem, while in
patient 1 it was delayed, of lesser intensity and shorter duration (patient 2, but not
patient 1, also showed modulation of nociceptive flexion reflexes RIII). No change
was seen in SI. Thalamic CBF changes were almost superimposable in both
patients, but pain relief was satisfactory only in one patient, in whom there was
also brainstem activation. CBF changes in OFC and anterior CC (ACC) were
stronger and more sustained in the patient with less pain relieving effect of MCS
than the other.

Pathophysiology: Human Data

8. Ness and colleagues (1998) studied a patient with paraplegia who, for many years,
experienced rapidly fluctuating, severe, highly aversive (VAS 10), unilateral pain
below the level of the lesion. The searing attacks lasted up to 10 seconds. SPECT
was done in pain and non-pain conditions (threshold of significance: 10%). When
experiencing pain, there was increased CBF to ACC (cingulate), increased thalamic
CBF bilaterally and increased SI contralaterally, plus decreased CBF in caudates
bilaterally. The patient responded to gabapentin, which reduced the anomalies (and
also induced mirror pain).
9. Doi and colleagues (1999) showed renormalization of thalamic SPECT hypoperfusion after successful convulsive therapy in 5 suprathalamic CP patients.
10. Fukui and colleagues (2002) reported thalamic hypoperfusion with
HMPAO SPECT in CPSP: ECT relieved the pain and renormalized thalamic
hypoperfusion.
11. Cahana and colleagues (2004) studied a patient with encephalitis and CP, who
showed left thalamic (Vc region) hypoperfusion. The patient complained of hot
left-sided paresthesias and burning pain, particularly in the chin and left palm,
plus evoked pains. SSEPs were normal. Lidocaine infusions relieved the pain and
the anomaly.
2. Studies assessing the evoked components
1. Cesaro and colleagues (1991) studied 4 CPSP patients with 123I-N-isopropyliodoamphetamine brain SPECT, with and without allodynic stimulation. In the two
patients with hyperpathia (with the lesions involving the parietal subcortical white
matter and the thalamocapsular area), there was hyperactivity (2026%) in the
central thalamic region opposite the painful side. Amitriptyline relieved both the
SPECT anomaly and the pain, and thermoalgesic deficits renormalized. The two
patients without SPECT anomalies had subcortical or subcortical plus thalamic
lesions.
2. We (Canavero et al. 1993, 1995) showed in CP patients that basal SPECT
hypoperfusion of SI increases under allodynic conditions and that this anomaly
spreads anteriorly to MI and other frontal areas.
3. The Lyon group (Peyron et al. 1998) studied 9 patients with acute unilateral CPSP
after a lateral medullary infarct (Wallenbergs syndrome) with PET (resolution:
7 mm). They did not study spontaneous pain (present in 4 at a VAS value of 35),
nor discussed baseline anomalies; brainstem and cerebellum were not studied.
All patients showed cold allodynia (assessed with frozen water in a moving flat
plastic container). During cold allodynia, statistically significant increases of
rCBF were seen contralaterally to stimuli in the lateral half of the thalamus, SI,
anterior insula and inferior frontal gyrus. R-CBF was increased bilaterally in SII
and inferior (opercular) parietal areas (BA3940) and significantly decreased
contralaterally in BA10, ipsilaterally in BA2432 and sub-significantly in ipsilateral
BA2331. No rCBF change was observed in BA24 (ACC). A significant decrease
was also seen bilaterally in BA1819. During electrically (high-frequency) elicited

243

244

Central Pain Syndrome

pain to the normal side, rCBF increased significantly bilaterally in BA3940 and
SII, contralaterally in BA6 (anterior insula), ipsilaterally in BA444547. rCBF
decreased ipsilaterally to stimulation in BA10. Again, no rCBF change was seen
in BA24. rCBF was significantly decreased bilaterally in BA1819. Cold stimuli
to the normal side induced significant increases in contralateral SII and BA3940,
without extending into SI, and ipsilaterally in BA46. No significant modification
was detected in the thalamus and ipsilateral parietal cortex. rCBF was significantly decreased bilaterally in BA1819 and ipsilaterally to stimulation in the
caudate head. There was a sub-significant decrease in contralateral BA2432 and
ipsilateral BA10.
The same group (Peyron et al. 1999) studied with PET 8 patients with CP (1 CCP,
3 brainstem CP, 1 thalamic CP, 3 corticosubcortical CP). They compared rest, cold
moving allodynia and thermal heat pain. They also studied 4 additional patients with
fMRI. Cold allodynia was associated with rCBF increases in contralateral insula-SII
and SI and bilaterally in posterior parietal cortex and ACC (plus ipsilateral
cerebellum). Thermal pain induced increased CBF bilaterally in insulae-SII, posterior
parietal, ACC and right prefrontal cortex (plus bilateral cerebellum), but not SI. MR
analysis showed individual variations in the allodynic response, except for the
contralateral insular-SII activity. Compared to thermal pain, allodynic pain induced
a greater activity in contralateral SI (ascribed to moving stimulus). Allodynic
pain compared to control stimulation of non-painful side showed higher activity
in contralateral SI and ACC.
They (Peyron et al. 2000) also reported on a CPSP patient, who complained
of spontaneous paroxysmal pain, mechanical and thermal allodynia and pinprick
hyperpathia. She had severe thermal hypesthesia of the left hand and foot. SSEPs
were diminished, but not absent. This case developed CP and allodynia in her left
side after a bifocal embolic infarct following vascular surgery involving both the right
parietal cortex (SI and SII) and the right rostral ACC (BA 24 and 32), plus
a small anterior and inferior part of the inferior parietal lobule (BA40), plus BA6, 8,
9 and 10. Judging from the images, SI could have still been partially active,
with reorganization posteriorly. SII was considered anterior to BA40 in the upper
bank of the sylvian fissure. This patient was studied with both PET and fMRI,
under basal conditions (PET only), control and allodynic stimulation. No rCBF
increase was found in any part of the residual cingulate cortices, neither in the basal
state (which included spontaneous pain and extensive hypoperfusion around the
infarct), nor during left cold allodynic pain (see their previous study). No
abnormality was observed in the left cingulate cortex. PET at rest (VAS 1) showed
a wide hypoperfusion including the infarct and widely extended within the frontal
and parietal cortices. Left parietal cortex, in the depth of SI, showed a significant
increase of rCBF in the control condition, which remained below the statistical
threshold for the allodynic condition. In the allodynic condition only, the rCBF
was significantly increased in the right anterior insula-SII, immediately forward
to (at the boundaries of the insular-SII infarct) the right parietal lesion; there
were also prominent responses in the hemisphere ipsilateral to allodynic (but
not control) stimulation: insula-SII and lateral thalamus and (sub-significantly) SI.
Sub-significant rCBF increases were observed in the head of the right caudate during

Pathophysiology: Human Data

the control condition and in the right lateral thalamus during allodynia. No rCBF
increases, even at a sub-significant threshold, were observed in ACC on both sides.
No intracerebral significant rCBF decrease was observed. Results remained
unchanged even on non-normalized PET images.
Finally, these authors (Peyron et al. 2004) studied the brain responses of
27 patients with peripheral (5), spinal (3), brainstem (4), thalamic (5), lenticular (5),
or cortical (5) lesions with fMRI, as innocuous mechanical stimuli were addressed
to either the allodynic territory or the homologous contralateral region. When
applied to the normal side, brush and cold rubbing stimuli activated the contralateral
primary (SI) and secondary (SII) somatosensory cortices and insular regions.
The same stimuli became severely painful when applied to the allodynic side and
activated contralateral SI/SII and insular cortices with, however, lesser activation
of the SII and insula. Increased activation volumes were found in contralateral
SI and primary motor cortex (MI). Whereas ipsilateral responses appeared very
small and restricted after control stimuli, they represented the most salient effect
of allodynia and were observed mainly in the ipsilateral parietal operculum (SII),
SI, and insula. Allodynic stimuli also recruited additional responses in motor/
premotor areas (MI, supplementary motor area), in regions involved in spatial
attention (posterior parietal cortices), and in regions linking attention and motor
control (mid-ACC).
4. Lorenz and colleagues (1998) studied a single patient who suffered Wallenbergs
syndrome with selectively abolished pain and temperature sensitivity in the right
leg. One year later, CP had developed in the leg, with touch and cold allodynia.
P40m dipoles calculated from magnetoencephalographic (MEG) fields after electrical
stimulation of both tibial nerves were localized in SI; however, stimulation of
the affected side caused deep pain sensations and elicited a large N80m component
best explained by an additionally co-active dipole in the cingulate cortex.
Cingulate activation was in the medial part slightly more posterior than BA24.
Electrophysiologically, the affected limb was characterized by larger components P40
and N80 of the tibial nerve SEP compared with the unaffected left limb. In particular,
the enhanced N80 amplitude augmented in parallel with the enhancement of CPSP
severity in the patient.
The same group (Kohlhoff et al. 1999) studied with MEG 4 patients with
Wallenbergs syndrome and CP. They found that the component around 80 ms
after tibialis stimulation showed side asymmetries in the patients which exceeded
the normal interindividual variability and were also reflected in the equivalent
current dipole parameters. The degree of asymmetry seemed to be related to the
severity of allodynia. They concluded for CP possibly reflecting functional
disorganization in SI.
5. Jensen and colleagues (1999) studied 10 CPSP women with H215O PET under
resting conditions and following stimulation of the painful body part and
the corresponding non-painful body part with phasic heat stimuli. They observed
hypoperfusion of the affected thalamic region versus non-affected thalamus under
resting conditions.

245

246

Central Pain Syndrome

6. Olausson and colleagues (2001) studied a hemispherectomized patient

with touch-evoked pricking and burning pain, plus a robust allodynia to brush
stroking (enhanced at a cold ambient temperature) in her paretic hand.
Psychophysical examination showed that, on her paretic side, she confused
cool and warm temperatures. On fMRI, brush-evoked allodynia activated posterior
ACC, SII and prefrontal cortex.
7. Morrow and Casey (2002) studied a CPSP man with H215O PET. He had sudden
onset of constant persistent painful dysesthesias of left hemisoma. Sensory
examination was normal, bar deep pressure allodynia on the left and elevated but
symmetrical cutaneous heat pain thresholds. MR disclosed a lacunar infarction
(2  4  7.5 mm) in Vc. At rest, rCBF was markedly reduced in the right Vc (as
compared to left Vc) and insula. Heat stimulation (4955) of either side showed
exaggerated rCBF increases relative to rest on the right (Vc and insula). They then
studied another 4 CP (CPSP, CCP) male patients (age: 4068), all with clinically
detectable impairment of heat and/or mechanical pain sensibility on the side of
CP. Each patient had abnormal, contralateral to pain, thalamic (3, hemithalamic
hypoactivity; 1, hemithalamic hyperactivity) and/or cortical asymmetry at rest and
increased thalamic and/or cortical responsiveness to contralateral stimulation
following contact heat stimuli.
8. Bowsher and colleagues (2004) studied four patients with small cortical infarcts:
one with a parietal operculum (SII) lesion, another with SII lesions encroaching on
the posterior insula, a third with damage to both banks of the sylvian fissure plus the
dorsal insula and the last with damage to the upper bank of sylvian fissure. In all, SI
was intact. Patients 1 and 2, but not 3 and 4, had (mild) spontaneous pain and also
pinprick and thermal anesthesia. In affected areas, mechanical pain was not felt in all.
FMR following thermal stimulation in patients 1 and 2 showed SII involved in
reception of innocuous and noxious thermal, mechanical and pinprick pains, and SI
in non-painful mechanical stimuli, although SI was activated in one patient by
innocuous cooling.
9. Seghier and colleagues (2005) studied a CPSP patient who suffered deep and
superficial burning cold-like constant and paroxysmal pain in the left hemisoma,
worse in the pectoral region, hand and foot. The pain was triggered by cold objects
and cool temperatures. He displayed a prominent mechanical allodynia. There were
severe left hypesthesia for heat, warm and cold temperatures, selective cold allodynia
and pinprick hyperpathia. On MRI, there was an infarct of Vc and adjacent internal
capsule (IC) along the STT. Under FMR conditions, the hand was stimulated with a
plastic object filled with water at 22, 15 and 5 (only 5 painful). Touch activated
bilateral SI, right SII and supplementary motor area (SMA). Increasing temperature
activated the right middle insula and right mesial SI. Hyperpathia activated BA24/32,
BA5/7 and the left anterior putamen. The activation in the putamen and BA5/7 was
ipsilateral to the stimulated hand. ACC activation was not correlated with the simple
cold quality of the pain-eliciting stimulus, as innocuous cold correlated with activity
in, for example, right insula and right SI.

Pathophysiology: Human Data

10. Villemure and colleagues (2006) reported on a patient with typical iatrogenic
cervical myelopathic CCP. The odor of cat litter, newspaper or popcorn triggered
electric shock/shooting paroxysms and also slowly increased spontaneous pain after
repeated challenge. Upon moving away from odors, pain abated. FMR under odor
challenge showed larger activations after the termination of the unpleasant odors
than after the termination of pleasant ones in the contralateral thalamus, amygdala,
insular cortex (bilaterally) and ACC, with similar trends in contralateral SI. Odors
triggered pain only on days they were judged unpleasant.
11. Ducreux and colleagues (2006) submitted 6 patients with syringomyelia
and suffering CP to fMRI. Cold allodynia (felt like a deep, freezing sensation,
sometimes burning, with a tingling sensation) under static conditions activated the
mid-posterior insula, ACC, SII, inferior parietal areas, frontal areas (BA8, 9, 45, 46),
mostly ipsilaterally and contralateral SMA. In 23 patients, activation in the
lenticular nucleus, hippocampus and cerebellar lobes was also observed. Brush
allodynia (felt like burning in 4 and electric shocks in 2) activated ipsi- and
contralateral SI-SII, inferior and superior parietal cortex, ipsi- and contralateral
middle frontal gyri (including BA 45-46), contralateral thalamus, caudate and SMA.
No activation was observed in BA24-32.
3. Studies assessing biochemical changes
1. Pattany and colleagues (2002) compared 7 SCI (plus 1 tumor) (1 C8, 6 T9L3)
pain patients with 9 SCI (plus 1 ischemia) (4 C48, 5 T7L3) non-pain patients and
10 controls in a magnetic resonance spectroscopy (MRS) study. A total of 74% of
pain patients had complete injuries (versus 67% without pain). Pain was described as
sharp, burning, aching or electric. Pain was generally above VAS 5 (86%). Statistical
analysis showed no significant differences in metabolite concentrations between the
two thalami. However, N-acetyl aspartate (NAA, a neuronal marker) correlated
negatively with average pain intensity and myo-inositol (a glial marker) correlated
positively. NAA also showed a significant difference between SCI patients with pain
and those without. Other trends toward significance remain of moot significance.
Limits of the study are inhomogeneity of ages between patient groups, exclusion
of females, scanning without drug wash-out and no differentiation between diffuse
versus end-zone pains.
2. Fukui and colleagues (2002) submitted to ECT a thalamic CPSP patient. 1H-MRS
(2  2  2 cm voxel in the thalamus bilaterally) was performed before and after
a single course of ECT. The NAA/Cr ratio was calculated. Before ECT, the L/R
thalamic ratio was 62.3%: after ECT (and during analgesia), the NAA/Cr ratio of the
left thalamus increased by 32%.
3. Willoch and colleagues (2004) reported on 5 right-handed CP patients (aged 54 to
77). In 3 cases, CP arose following an ischemic stroke also involving the thalamus,
in 2 after a hemorrhagic stroke (pons; parietal angioma). Both spontaneous and
evoked components were present in all and involved the hemibody, bar the face
in one. CP never started immediately after the insult. They assessed diprenorphine
(DPN) binding with PET. Arterial sampling necessary for quantitative modeling

247

248

Central Pain Syndrome

could not be performed in 3 patients. Results were compared with 12 healthy

controls with a mean age of 39 years. Given low opioid receptor (OR) binding, SI was
excluded from the analysis. This disclosed a hemispheric asymmetry with significant
relative reductions in OR binding in prefrontal BA44, parietal BA40, SII and insula
(BA14) and Vc contralateral to symptoms. The insular cluster was adjacent to SII and
probabilistically extended into SII. While Vc showed maximal peak difference, there
was reduced binding also in anteromedial thalamic nuclei. A bilateral relative
reduction in OR binding was shown along the midline in the ACC (BA24 and 32),
PCC (BA7 and 31) and the PVG. The ACC revealed maximal reduction posteriorly,
but stretched to BA24 and 32. Non-significant reduced OR binding was observed
in BA6/8 and BA21/22/38. There were only reductions compared to controls and no increases.
Actually, infarcts in the thalamus and parietal
cortex could have been at the basis of the observed
reductions. All 3 patients with thalamic lesions
demonstrated binding levels below the control
group, but the two patients with cortical or
pontine lesions revealed reductions in the lowest
range of the patient group. The global mean value
of DPN binding for 2 patients was within normal
range as compared to the control group.

Figure 7.2: Iodine-123-labeled iomazenil SPECT in

one BCP case. Data processing: one week after
suspension of all drugs, and following thyroid block
with oral potassium perchlorate, SPECT scanning
started 70 minutes after intravenous injection of
iodine-123-labeled iomazenil (MallinCkrodt),
111 MBq, with a rotating two-head SPECT device
(Varicam, General Electric equipped with highresolution, low-energy collimators). SPECT images
(120 128  128 pixel matrices, 5 pixel thick
slices, zoom 1, 30 angular 30-second-long steps;
160 keV 20% window recording at least 3500
kCnts) were reconstructed from projection data by a
filtered backprojection technique with a Hanning
filter (cutoff frequency 19 f/Nyquist, power factor
40). Cortical regions of interest were automatically
marked by the dedicated GE reconstruction software on operator-chosen transaxial slices.
Semiquantitative right-to-left ratios were immediately calculated by the package software with a
+10% significance limit. Arrowheads show the
anomalous frontoparietal area.

4. Jones and colleagues (2004) performed a

similar study in a group of CP patients (predominantly CPSP) compared to age-matched pain-free
controls. They observed reductions in opioid
receptor binding mainly in the dorsolateral (BA
10) and anterior cingulate (BA 24, with some
extension into BA 23), plus insula and thalamus.
There were also reductions in the lateral pain
system within the inferior parietal cortex (BA 40).
These changes in binding were outside CT/MRI
areas of damage.
5. By means of iodine-123-labeled iomazenil
SPECT, we assessed the regional distribution of
benzodiazepine-GABA A receptors in the cortex in
five patients with CP (three women and two men;
aged 4165; time from onset: at least three years;
three patients with a neuroradiologically confirmed thalamic and/or capsular previous stroke,
ischemic or hemorrhagic and two with pure spinal
cord damage due to previous myelitis and no endzone pain). Four patients showed reduced uptake
at parietal and, in two cases, frontal cortical levels
on the side opposite the painful syndrome (R/
L 117, 116, 113, 114). In the fifth patient

Pathophysiology: Human Data

(a thalamocapsular hemorrhage), the ratio approached significance (R/L 0.91). Both

brain and cord cases displayed similar binding anomalies, with reductions in CCP
contralateral to worse pain, excluding direct brain damage of GABA receptors as
a mechanism of such reduction (Figure 7.2).
interpretation: Neurometabolic studies suffer from serious drawbacks (Box 7.1).
Thus, only general conclusions are possible regarding the genesis of CP:
1)
2)
3)

4)

5)
6)

The thalamus appears to be implicated.

Somatosensory areas (SI-SII) appear to be involved.
The ACC cannot be a prime actor: rCBF changes in ACC are also reported
in PNP and other chronic pains (Hsieh et al. 1995), making it an unspecific
finding, and a lesion of ACC does not prevent or is involved in the generation
of CP. Allodynia does not necessarily activate ACC (as in studies of acute
pain in healthy volunteers). Absence of change in ACC (plus SI-SII) has been
reported for capsaicin allodynia as well (Baron et al. 1999).
Bilateral activation of brain areas in CP is possible simultaneously: normal
inhibitory mechanisms cannot rein in incoming impulses, with spread of
(de)activations (Box 7.2).
All CBF changes are functional and rapidly reversible, rejecting entrenched
neuroplasticity theories (see Chapter 8).
Allodynia involves different changes from spontaneous pain: spread to frontal
areas may signal engagement of avoidance networks, as unpleasantness degree
increases.

Based on such evidence, several older and newer theories of CP collapse, as they
ignore the role of cortex or ascribe the pain generator to a dedicated CNS area
outside the thalamocortical sensory loop (see below).
Reduced transmitter binding may be due to direct neuronal damage, anteroor retrograde transneuronal degeneration, release of endogenous molecules and
subsequent increased occupation of binding sites, internalization and/or receptor
downregulation. In this regard, the study of Willoch and colleagues does not provide
useful data. Decreased opioid binding (i.e., increased production of endorphins) is
seen in many pain conditions. Aside from suboptimal receptor binding specificity
of available markers and poor CP responsiveness to opioids, we never observed CP
worsening during opioid challenge, and naloxone has been shown to be ineffective in
the only controlled trial (see Chapter 5), refuting an opioid hypertonus as surmised
by these authors. In this context, Morley and colleagues (1991) found reduced spinal
(dorsal horn) enkephalin concentrations at segmental levels corresponding to
the pain, at sites where primary sensory afferents terminate, in two cases of chronic
pain. Proper control, i.e., non-CP patients with similar lesions, to study patients
and homogeneous age (not a group of younger healthy subjects) was lacking. Finally,
they erroneously compared their resting findings with imaging studies of CP during
allodynic stimulation, two very different situations. A literature survey points
to similar brain areas activated by evoked pain in both PNP and CP, making these
findings unspecific. A simple corollary endorphin fall-out due to the primary
lesions can explain their findings (along these lines, we might also expect possible

249

250

Central Pain Syndrome

Box 7.1 Limits of neuroimaging studies
1. Very large discrepancy between actual decreases in spiking activity and rCBF decreases (by a factor
3 to 7): under pathologic conditions neuroimaging methods based on hemodynamic signals may
only show small changes, although the underlying decrease in neuronal activity is much larger (Gold
and Lauritzen 2002).
2. Complex mechanisms of cortical activation, even in cases of simple sensory stimulation:
dissociations may occur between obvious neurological deficits and apparently normal activation
patterns, i.e., activation studies should be interpreted cautiously in patients with focal brain lesions
(Remy et al. 1999). Also, noxious stimuli produce arousal, orientation, escape or immobilization
and help-seeking, which must be properly dissected.
3. False negatives due to arbitrary group analyses (averaging) that miss important individual rCBF
changes due to intersubject variability (e.g., anatomical [cingulate gyry patterns, thalamic/pallidal
size and location of tactile representation of various body regions somatotopy differ among
subjects], attentional [requirement of attentional resources is longer after pain than other sensory
stimuli], affective [e.g., anxiety], previous experience) or averaging over task duration. Various
combinations of cortical and/or thalamic activations, uni- or bilateral, in individual subjects
following the same stimulus and a high degree of variability in cortical activation patterns seem the
norm, calling for repetitive single-subject analysis, particularly during heat- and cold-evoked pain.
Intensity of a stimulus may be rated similarly among subjects, but overall sensory-cognitive
experience of that stimulus may vary. Comparisons of different individuals have shown that fMR
responses to the same stimulus within a particular area are variable in location and can also differ in
extent over time depending on pain duration and intensity. Hence the importance of single case
studies.
4. Different neural recruitment depending on methodology of noxious/non-noxious stimulation (contact
versus noncontact, escapable versus nonescapable pain), type (cold pain, more unpleasant versus
heat pain, less so, having only some regions in common; tonic cold pain and phasic heat pains
having different sets of afferents and different sympathetic activation; pinprick engaging the lateral
system, heat pain both lateral and medial pain systems), duration (tonic stimuli being more
unpleasant than acute ones at any given level of pain intensity), location (skin, subcutis, muscle:
frontal areas and SII more activated with skin than muscle pain), periosteum, vascular nociceptors
and side of the body), quality and, most importantly, intensity (which bears on attention, arousal,
orientation and intrusiveness) of stimulus (Bushnell et al. 1999). Responses are usually lateralized
and most often contralateral to a noxious stimulus, but the side can be unpredictable. Different
operational mechanisms recruited in processing a long-lasting pain state with persisting emotional
distress versus acute pain with a low affective tone.
5. Low sensitivity of fMR to small, but important, differences in cortical activation, to cold pain and to
deep structures.
6. Different spatial resolutions among studies (due to different techniques and different generations of
machines) fMR 4 PET 4 SPECT which cannot yet resolve, for example, single thalamic nuclei or
SII from insula, and may wash out small foci of activation during averaging over a wide neuronal
population.
7. Different temporal resolution of chosen technique (PET/SPECT versus fMR versus MEG/EEG), with
brief, transient changes or frequent fluctuations in neuronal activity, which may be a critical
component of the brain process under investigation, going undetected: non-converted BOLD fMRI
cannot capture the initial highly localized increase in O2 consumption (proportional to initial
changes in neuronal spike frequency) following a stimulus, with serious mapping consequences
(Smith et al. 2002); dissociation between changes in synchronization of neuronal populations and
(no significant) changes in mean neuronal firing rates (mute imaging).

Pathophysiology: Human Data

8. Different values of significance in rCBF changes (generally 35%).
9. Dependence of PET on a nitric oxide mechanism which is not evenly distributed (making absence
of CBF change not equivalent to absence of activity) and unclear mechanisms of coupling of
glucose and O2 consumption to brain activity.
10. Inability of SPECT/PET to distinguish between increased inhibitory and excitatory activity.
11. Non-quantitative nature of fMR versus quantitative analysis (including of basal state) possible with
SPECT/PET.
12. Poor slice selection and low signal-to-noise ratio.
13. Inability of fMR to scan the whole brain (unlike SPECT-PET) and the basal resting pattern of
activation.
14. Widely different statistical factors among studies (number of patients, values of significance,
correction for multiple comparisons versus no correction, wrong selection of control group,
especially in CPSP study, controls being often younger), making comparison impossible; small
differences in setting up superficially similar experimental tasks leading to markedly different
neuroimaging results.
15. Nonhomogeneous degree of differentiation of sets of increasingly intense stimuli analyzed with
subtraction analysis (e.g., subtraction between neutral and more intense, but non-painful, heat
and neutral and painful heat yielding differences of greater magnitude than subtraction between
intense, but still nonpainful, heat and painful heat), which calls for simultaneous correlation
analysis.
16. Different data acquisition and analysis procedures between, for example, SPECT and PET.
17. Artefacts (e.g., bilateral increases in temporal muscle blood flow mistaken for brain activation;
geometric distortion (MR techniques), interference with resolution from large veins (fMR), imperfect
correspondence between fMR signal and locus of synaptic activity).
18. Unexplained participation of areas not believed to be involved in a task confounding interpretation
(e.g., bilateral visual areas decreases).
19. Disregard of the high baseline activity in the awake resting brain; most functional imaging
experiments show small fractional changes in CMR-O2 from baseline values in response to
stimulation, but not the larger increases in the overall cerebral metabolic rate of glucose
consumption (i.e., uncoupling between utilization of glucose and oxygen). Importantly, such
baseline activity might be high enough so as not to require incremental activity during
performance, i.e., a particular region could still be actively contributing to brain function. Neuronal
activity in the cortex is extremely efficient, with neurons requiring a minimum total amount of
energy to process information. This would call for studying the magnitude of the total neuronal
activity (baseline plus activation) using measurements of neuroenergetics (Shulman et al. 2004).
20. PET estimated CBF changes identified using the relative CBF analysis not necessarily reflecting
functional change, particularly when the experimental conditions directly affect global CBF
(absolute CBF analysis should be considered when conditions potentially evoke autonomic nervous
system responses).

deranged NE/5HT neurotransmission). The same comments apply to the study of

Jones and colleagues.
In our own study, reduced binding could have been due to decreased sensitivity
of the postsynaptic membrane to GABA or a downregulation of receptors in response
to enhanced release of GABA. A significant GABA A downregulation, in the course
of long-standing CP, at fronto(MI/premotor/PFC)-parietal (SI) level (and not
diffusely) is suggested by our data.

251

252

Central Pain Syndrome

Box 7.2 Explaining deactivations
While hyperactivations imaged by current technologies can be more easily explained, deactivations have
not yet been adequately explained. In particular, noxious input may initially activate, and after some
time depress certain brain regions. Backonja and colleagues (1991) suggested that initial somatosensory activation in response to tonic pain (decreased alpha 1 power) progresses to somatosensory
inhibition (alpha 1 augmentation) after the first minute of stimulation. Le Pera and colleagues (2000)
showed that tonic muscle pain induces EEG increments of both delta and alpha 1 powers bilaterally
over the parietal somatosensory areas (but not SII-insula or ACC), more so contralaterally to painful
stimulation. Since enhanced slow waves are usually considered an expression of inhibition, these
findings could be related to inhibitory processes occurring in SI. Apkarian and colleagues (1992)
showed SI deactivation following contralateral stimulation with moderately painful hot water bath.
A SPECT study found that thalamic perfusion increases just after onset of symptoms as a reaction to
pain and then gradually decreases in more chronic phases (Fukumoto et al. 1999). High intensity
stimuli produce SI deactivation. Thus, tonic pain may trigger an inhibitory response in these areas.
The same line of reasoning applies to CP.
Possible mechanisms include the following. (1) Organized, baseline default mode of brain function,
which is suspended during activating contexts (Raichle et al. 2001) or pathologic states. (2) Ongoing
inhibition (Canavero et al. 1993): activity of GABA neurons demands energy (i.e., enhanced glucose
metabolism; Ackermann et al. 1984), but the net inhibition may swamp their increased demand for
energy. Also, few interneurons (whose firing rate tends to be higher than pyramidal neurons) can
effectively inhibit many projection cells, particularly if GABA is over-released by these or GABA receptors
are increased on target projection neurons. (3) Diminished input: this cannot be the explanation in the
CP setting, since excitatory allodynic barrage is accompanied by further deactivation. Also, thalamic
hypometabolism renormalizes, along with analgesia, after cordotomy (i.e., further input reduction)
in cancer patients (Di Piero et al. 1991). (4) Passive shunting to nearby activated areas. (5) Decrease
in thalamic firing between bursts (Lenz 1991), due to excessive inhibition in the thalamus trying
to overcompensate excessive excitatory nociceptive input (this cannot be the case: see section on
neurophysiology and Box 7.3). (6) Attentional focusing on an area and shutting down of another;
even anticipation of a painful stimulus yields decreases in blood flow in areas of SI outside the
representation of the anticipated stimulus (Drevets et al. 1995). (7) Diminution of ongoing neuronal
processes as an outcome of increased neuronal activity elsewhere. (8) Diaschisis (Nguyen and Botez
1998), defined as a sudden inhibition of function produced by an acute or chronic focal disturbance in
an anatomically intact portion of the brain remote from the original site of injury, but anatomically
connected with it through fiber tracts. There are several forms: (a) the transhemispheric form
acts through the corpus callosum and may be due to loss of facilitatory inputs; in the acute phases
of stroke there may be reactive contralateral disinhibition or facilitation followed by depression; (b) the
corticothalamic form, which follows pure cortical stroke, is accompanied by ipsilateral thalamic
hypometabolism (e.g., Kuhl et al. 1980: 5 non-CP stroke patients); (c) the thalamocortical form, in
which both small and large posterior thalamic infarcts can result in ipsilateral parietotemporal
hypometabolism; unilateral thalamic stroke may induce bilateral (ipsilateral 4 contralateral) metabolic
cortical depression, perhaps mediated by the corpus callosum; and (d) reverse diaschisis (i.e., increase
in CRF) determined by contralateral structures (Weiller et al. 1992) (interestingly, isolated lesions of the
internal capsule show no significant cortical hypometabolism). (9) Neuronal death. This has been
excluded (Baron et al. 1986): cortical hypometabolism shows a trend toward renormalization over time.
Most importantly in the CP setting, both pain and (de)activations can be promptly reversed.
Among these, inhibition is a prime candidate. For instance, different net effects of excitation and
inhibition have been observed within SI, with nociceptive neurons even suppressed by noxious stimuli
(see references in Bushnell et al. 1999; Schnitzler and Ploner 2000). Inhibitory effects within SI have
also been demonstrated simultaneously with (Tommerdahl et al. 1996, 1999) and after excitation
(Backonja et al. 1991), within (Tommerdahl et al. 1996, 1998) as well as outside the somatotopically
appropriate regions of SI (Derbyshire et al. 1997). Inhibition within SI is known to enhance contrast
both within-area and with contralateral SI for pain perception (Drevets et al. 1995). The net effect of

Pathophysiology: Human Data

exciting some neurons and inhibiting the spontaneous activity of others could have different effects on
PET rCBF or on fMR measured venous blood oxygenation. Recently, a TMS study found that pure sensory
thalamic stroke, which reduces or abolishes sensory input induces a hyperactivity of inhibitory cortical
neurons and simultaneously induces intracortical excitability without affecting corticospinal excitability
(Liepert et al. 2005).
Thalamic hypoperfusion has been reported not only in CP, but also in PNP and cancer pain (Di Piero
et al. 1991; Hsieh et al. 1995), making it an unspecific finding. However, the underlying mechanism
may be different between CP and other pains, as propofol only relieves CP, along with this anomaly
(Canavero et al. 1995). Part of the normal tonic synaptic thalamic activity may be concerned with
inhibition of pain perception/input and this may be defective in CP, the thalamus being hyperresponsive
following innocuous or noxious stimuli (allodynia-hyperalgesia).
By focusing on the pharmacodynamic profile of propofol (Chapter 5), we may speculate on the origin
of such deactivations. One possible mechanism would be disinhibition, that is, CP would be subtended
by ongoing hyperinhibition at cortical and/or thalamic levels: our binding study may be explained by
both reduction of GABA receptors and GABA hypertonus displacing the tracer. We know that, under
normal conditions, there is tonic inhibition at both thalamic and SI levels and for some reason this
would be increased in the CP setting. On the other hand, almost half of patients do not respond to
propofol and we speculated that this might depend on too strong an excitatory tone in the cortex
(Canavero et al. 1996). Since the end result in CP must be net corticothalamic facilitation (Chapter 8),
this inhibition would be sufficient to produce deactivation, but not switching off descending input, which
seems indefensible. In fact, ketamine can quench CP by antagonizing intracortical excitation and some
studies actually point to cortical excitation and thalamic deactivation, so that one possible explanation
would rest in study methodology (Hirato et al. above versus our data). Moreover, some patients with
thalamic deactivation showed no anomaly at SPECT. Differential participation of separate cortical layers
too may originate different findings on SPECT or PET. Feedback excitatory connections coming from a
higher order cortical area densely project to layer I, where they may activate pyramidal neurons by
synapsing on their apical dendrites. In parallel they might also activate GABAergic interneurons located
in this layer and in turn may inhibit the same pyramidal cells, so that inhibition of an area may disinhibit
another. This might explain hyper- and hypoactivations of different areas. In fact, GABA inhibition
reaches both horizontally through long-range monosynaptic projections (surround inhibition) and
vertically in the same column (vertical inhibition): pyramidal cells found in layers VVI are under stronger
inhibition than those in layers IIIV (Shepherd 2004). Inhibition, besides adapting receptive fields (RFs)
of pyramidal cells to context, has also a synchronizing role; interlaminar inhibition, for one, has an
important role: the synchronous activity of even a small number of inhibitory cells (unlike pyramidal
cells) making many contacts onto postsynaptic cells could be sufficient to provide synchronization in a
large population of pyramidal neurons. Specularly, the synchronous activation of a local group of many
pyramidal neurons may provide an optimal stimulus for activation of inhibitory neurons, compensating
for the sparse connectivity from individual pyramidal neurons onto interneurons. The extensive recurrent
excitatory connections between pyramidal neurons allow positive feedback to dramatically amplify
afferent signals, important in enhancing cortical sensory selectivity. However, these circuits are
intrinsically unstable. This is kept in check by GABAergic interneurons, in a tightly regulated balance.
Thus, it is not difficult to envision different degrees of activation and inhibition (even overinhibition
leading to bursting: see Box 7.3), simultaneously or not, even in the same lamina. To this, we must add
regional differences in GABA and glutamate release (Salin and Prince 1996; Castro-Alamancos and
Connors 1997; Shepherd 2004). An imbalance in excitatory and inhibitory influences most likely will
not consist of uniformly increased excitation and decreased inhibition. Spatial and temporal changes
are the norm. Under physiological conditions, the shifting balance between these components serves
the scope of promoting contrast enhancement to improve discrimination or curbing strong inputs
(centersurround interactions), primarily in supragranular layers. The check is a shift over time that
favors inhibition (i.e., prolonged firing induces a much stronger depression of pyramidal excitatory
synapses than of interneuronal inhibitory ones, never vice versa), particularly at relatively high
frequencies (Galaretta and Hestrin 1998; Nelson and Turrigiano 1998), or during prolonged sensory
stimulation (such as during chronic pain). This high-frequency shift may be due to sensory adaptation of

253

254

Central Pain Syndrome

BOX 7.2 Explaining deactivations (continued)

excitatory neurons at lower frequencies than inhibitory neurons, increase of excitatory inputs to
inhibitory neurons at higher frequencies and depression of excitatory inputs to pyramidal neurons in SI
or feed-forward inhibition predominating at lower amplitude input (Moore et al. 1999). At shorter time
scales, excitatory inputs to some classes of interneurons show transient facilitation, promoting stability
by boosting recurrent inhibition.
Thus, a general model for theoretical discussion emerges (Canavero et al. 1996; also based on
Thomson and Deuchars 1994).The direct monosynaptic thalamocortical (TC) input and pyramidal
neuroninterneuron inputs involves non-NMDA receptors, unlike local circuit pyramidalpyramidal cell
connections. These connections, which act at distal dendritic sites, can easily trigger a reverberant
excitatory activity between interconnected pyramidal cells and recruit surrounding columns when they
receive coincident TC afferent input. Random tonic activity will more readily recruit other pyramidal
neurons, while burst firing will recruit interneurons. This simple circuit favors excitation and recruitment
of surrounding columns when excitatory inputs to pyramidal neurons are weak or desynchronized,
particularly when the input is repetitively activated. In contrast, it favors inhibition when inputs
are asynchronous or strong, resulting in a strong inhibitory surround, limiting reverberant excitation
amongst pyramidal neurons. In other words, when many columns are recruited for a long time, given the
presence of powerful lateral inhibitory circuits, reciprocal inhibition between columns might exert an
increasingly dominant role. The strongest excitatory connections between pyramidal neurons appear
to involve connections between neurons within a column or very closely neighboring columns, which can
recruit each other in a reverberant manner, i.e., vertically excitatory connections appear to predominate.
In contrast, inhibitory circuits appear predominantly to involve lateral connections. Inhibitory interneurons with vertically oriented axons innervating cells within their own column might therefore receive
excitation from surrounding columns. Conversely, an interneuron excited by pyramidal neurons in its own
column would inhibit surrounding columns. This model must also take into account that NMDA receptor
density and subtypes vary greatly between cortical areas and there may even be a decrease of NMDA
activation with age (Castro-Alamancos and Connors 1997). Only further studies will determine the
applicability of this model to CP.

NEUROPHYSIOLOGY
1. Human microrecording/stimulation studies
Findings in BCP
1. Obrador and colleagues (1957) failed to elicit pain by stimulating the thalamus
in cases of CP.

2. Nashold and Wilson (1970) reported on 3 CP patients. One (V.H., female) was
affected by severe paroxysms of right lancinating facial pain plus dull, aching pain
(thalamic pain), both worse in the cheek (which became red), due to vascular
mesencephalic lesion associated with subarachnoid hemorrhage. During the
pain paroxysms, EEG recording demonstrated in the left dorsal mesencephalic
tegmentum epileptiform spike activity grouped in trains lasting for the duration of
the pain, and less striking EEG spikes coincident with dull aching pain. Electrical
stimulation of this area enhanced the paroxysms and a radiofrequency lesion
eliminated both the abnormal EEG activity and the pain. Interestingly, despite gross
anomalies in the anterior parietal lobe and left frontoparietal white matter (single
spikes or multiple bursts at 6 s and beta rhythms mixed with spike activity, with
bursts every 13 s, minimal beta activity and slow theta), stimulation at these sites

Pathophysiology: Human Data

elicited no subjective responses. A second patient (P.B., male) suffered burning CP

to right face, arm and chest due to a traumatic parietal and stereotaxic midbrain
lesion. Four lesions in the left dorsolateral mesencephalon, in the region in which
stimulation reproduced the pain, relieved both the pain and the hyperalgesia,
although an undefined discomfort in his hand lingered on. Two years later, he
suddenly died from acute subdural hematoma. At autopsy, an atrophic lesion was
found in the left parietal lobe. A third patient (S.M., female) suffered burning/
freezing CP to the right hemisoma following thalamomesencephalic stroke. Two
lesions were made in the left dorsolateral tegmentum where stimulation elicited
the pain; pulvinar stimulation was silent. The patient still felt the cold sensation
in the arm, but it was no more unpleasant.
3. Guecer and colleagues (1978) implanted electrodes stereotactically in (likely) Vc
and nearby somatosensory nuclei and made thalamic EEG recordings (scalp EEG plus
thalamograms) in 7 patients with thalamic CPSP. Excessive thalamic slowing was
found in 4/7 (3 within range). One patient had marked rhythmical intermittent delta
activity in the thalamus which was often triggered by arousing stimuli. Thalamic
spindle activity was sometimes noted without concomitant spindle activity on the
scalp and would occasionally occur in states of early drowsiness. All 3 patients with
markedly abnormal scalp EEG recordings also showed excessive slowing in the
thalamic leads. Marked thalamic and surface slowing of irregular (polymorphic)
waveform was found to increase in the thalamic as well as the scalp leads when the
patient became drowsy. In 2 of these 3 old patients, abnormal EEG scalp findings
were likely due to advanced diffuse cerebrovascular disorder. Thalamic participation
in the posterior alpha rhythm was absent or poorly developed in most patients: only
2 had good evidence of alpha rhythm, possibly depending on the electrode site (and
on the degree of cortical alpha development). Marked thalamic delta activity likely
marked a genuine pain-related abnormality (insertion trauma was ruled out by
concomitant scalp EEG slowing and lack of subjective implantation complaints).
4. Namba and colleagues (1984) reported on 11 patients with BCP. Stimulation
in mesencephalic lateral tegmental field elicited the most severe burning
pain compared with Vc and internal capsule.
5. Barcia-Salorio and colleagues (1987) studied 2 patients with CPSP. The
preoperative EEG of patient 1 showed basal activity and marked bilateral
delta waves, worse on the affected stroke side in temporal regions. The second
showed slow irritative activity on scalp EEG. Deep brain recordings of scalp EEG
showed marked delta activity in the thalamus of patient 1 and a cortical focus in the
second case. After radiosurgical Vc thalamotomy, these findings were unchanged,
despite some pain improvement.
6. Ohyes group (Hirato et al. 1991), in a series of 11 patients with BCP (plus 5
Parkinsons disease controls), noted that deep pain was more marked in nonthalamic lesion (on CT) cases and superficial pain in cases with definite thalamic
damage. Patients were submitted to microelectrode recording. In the non-thalamic
lesion group with CP, the power amplitude voltage histogram showed a slight
reduction with a mixture of various activities in and around the Vim nucleus

255

256

Central Pain Syndrome

and multiple peak configurations between 0 and 1000 Hz with a maximum at 200
300 Hz. Thalamic background neural activity in and around Vim was comparable to
controls. Background neural activity in intralaminar nuclei (CL) was generally low.
In thalamic CP, the power amplitude voltage histogram (i.e., background neural
activity) showed marked decrease in and around the Vim nucleus (which shows
clusters of STT fibers), suggesting damage in Vc. The background neural activity in
CL was higher than in Vim, especially in its dorsal part, and was also higher than in
the non-thalamic lesion group. In a case without any CT lesion, but a dominant
superficial pain, the background neural activity in CL was relatively high. Thus, in
non-thalamic CP (deep pain dominant) thalamic background neural activity was
relatively high in Vim (where deep muscle sensation can be usually elicited), but low
in CL, whereas in thalamic CP (superficial pain dominant), this was higher in CL than
in Vim and markedly decreased in Vc. The initial small damage in Vc may have
induced an abnormal state of activity in the surrounding areas in surviving Vc
neurons and adjacent Vim neurons (and their projection areas). Ohye (1998) very
often found that spontaneous activity in Vim and Vc of CP patients was considerably
reduced, particularly with massive thalamic involvement. Many irregular burst
discharges were encountered throughout the electrode descending in these nuclei, but
he noted no coincidence between pain sensation and moment of burst discharge.
The topographic representation in Vim and Vc was lost. He also found more
responses related, for example, to face and arm and often convergent responses from
different peripheral receptive fields (RFs). Moreover, a response to ipsilateral stimuli
was found. Neurons of the face area (including eye movement neurons) seemed to
occupy a wide area of Vim. Curiously, coagulation in this area did not change eye
movements, but relieved deep pain.
7. Fukaya and colleagues (2003) reported on cortical stimulation findings in
31 CPSP patients (28 thalamo-putamino-capsular; 3 Wallenbergs syndrome). In 23
(84%), SI stimulation at 50 Hz elicited contralateral tingling versus 40% of non-pain
patients; in 12 (39%), abnormal pain sensation or exacerbation of original CP were
observed, versus 0% of non-pain patients. MI stimulation at 50 Hz had no motor
effects, but evoked sensory tingling in 52% of the patients versus 20% of non-pain
patients, and very unpleasant sensations (interpreted as a sign of extensive
reorganization and unfavorable prognostic sign for MCS-induced analgesia) in
6% of the patients, versus none of non-CPSP cases. MI stimulation at 12 Hz evoked
tingling in 25% of the patients. In these authors experience, half of their CPSP cases
submitted to Vc DBS reported more pain.

Findings in CCP

spinal recordings:
1. Loeser and colleagues (1968) recorded unit activity in the dorsal horn of
a chronically denervated conus medullaris of a paraplegic suffering from burning
rectal and thigh pain and hyperpathia following trauma: denervated cell groups
(10 dorsal horn neurons rostral to the site of injury) had developed spontaneous
high-frequency epileptic paroxysmal burst discharges.

Pathophysiology: Human Data

2. Evidence of high-level spontaneous activity assumed to be abnormal focal

hyperactivity within the superficial laminae of the injured cord has been recorded up
to 7 levels cephalad to injury site prior to computer-assisted DREZ surgery for SCI
and other pains (39% of cases had hyperactivity higher than 3 levels above injury
site) (Edgar et al. 1993).
3. Falci and colleagues (2002) performed multilevel DREZ surgery on 41 CCP
patients. Electrophysiological analyses of the DREZs were performed one level caudal
to the injury site and up to 5 DREZ levels cephalad, exploiting an active electrode
inserted free-hand 2 mm into the specific DREZ tilted 3545 medially (the same
as per coagulation). In 32 patients, additional DREZ recordings were carried out
during transcutaneous C-fiber (inclusive of sympathetic fibers) stimulation in which
a current perception threshold device was used (electrodes were in the distribution
of a dermatome, with 5 Hz electrical stimuli activating the nerve fibers directly, but
not the actual receptors in the skin due to too low current levels). The device was
used for preoperative testing of dermatomal skin sensation in a C-fiber frequency
band caudal to, at, and cephalad to injury level. A 5 Hz threshold above 0.35 mA was
empirically assumed as significant. In general, the elevated thresholds were found
in dermatomes at and cephalad to the neurological injury level in patients who
were sensory complete (occasionally also in dermatomes immediately caudal to the
sensory-complete neurological level); these same skin dermatomes with elevated and
presumed abnormal thresholds received above-threshold stimulation intraoperatively.
Intramedullary recordings were then made in the DREZs corresponding to the
particular skin dermatome. Data were analyzed and filtered to obtain spindles,
presumed to signal abnormal neural activity when exceeding 3 s. These were
corroborated by higher voltage and frequencies of the activities. The same recordings
were obtained after lesioning. These data were in spatial correlation with those
obtained with current perception threshold. In the first 9 patients, 7 showed areas
of DREZ neuroelectrical hyperactivity: radiofrequency microcoagulations (90C for
30 s) with 1 mm of separation were performed in order to silence all abnormal
activity (otherwise, they were repeated). In the 2 cases without hyperactivity,
lesioning extended at 2 DREZs cephalad to injury level and 1 below (90 for 30 s).
Of the remaining 32 patients, 9, all with below-level pain, had no spontaneous
DREZ hyperactivity; operative transcutaneous C-fiber stimulation of skin dermatomes with elevated C-fiber sensory thresholds resulted in evoked neuroelectrical
hyperactivity in specific DREZs, presumed pain generators, and used to guide
lesioning: 8 were totally relieved, with 1 failure. In the rest, both techniques guided
total silencing of hyperactivity (see results in Table 7.1, below). Lack of spontaneous
neuroelectrical hyperactivity in 27% of the patients was ascribed to pain being
cyclical and waxing and waning in intensity.
cerebral recordings: Lenz (1991 and references therein; Lenz et al. 1994)
studied patients with CP following spinal cord transection. All patients experienced
pain in the anesthetic part of the body; some also experienced dysesthesias in the part
of the body adjacent to the area of sensory loss. They designated the area of thalamus
representing the borderzone area and the anesthetic area as the borderzone/anesthetic

257

258

Central Pain Syndrome

area (BAA). Evidence of somatotopic reorganization was found. Neurons with

RFs on the border of the area of sensory loss occupied more of the thalamic
homunculus in Vc than in patients with controls (movement disorder patients),
i.e., body parts bordering the anesthetic body part had increased representation.
For instance, in one patient, the representation of the trunk occupied 1.2 mm of
a trajectory through the part of the thalamus where the leg, anesthetic as a result of
the spinal injury, is often represented. In another with clinically complete spinal
transection at C6, the representation of the external ear, neck and occiput occupied
1.5 mm of a trajectory through the forearm representation, versus 0.10.3 mm of
neck and trunk representation in controls. Stimulation of these neurons by whatever
means (e.g., touching the skin near the border of the sensory loss) could produce an
abnormal sensation in the anesthetic part of the body (mislocalization). A significant
increase in the number of neurons in Vc (BAA) without RFs was also characteristic.
Unlike controls, Vc microstimulation at sites with neuronal RFs on the border of
the anesthetic area of the body characteristically revealed a dissociation between the
RFs and projected fields (PFs) (RF/PF mismatch), with PF altered less than the
somatotopic map of the inputs demonstrated by the RF; RFs were often located
on the border of the anesthetic area, while PFs extended far into the anesthetic
part of the body, suggesting to the authors that abnormal activity recorded in
borderline regions might be reflected in sensations experienced in anesthetic areas,
but also that the representation of sensory input (RFs) is much more plastic than
the central representation of the part of the body (PFs). In other words, in Vc
regions that would normally represent the anesthetic body part, neurons often
had no RFs, although PFs were referred to the anesthetic body part, evidence that a
central representation of the anesthetic body part still exists years after total interruption
of input from that part of the body, an essential ingredient if pain is to be appreciated
in that body part. Microstimulation at these Vc borderline regions often produced
sensations in the anesthetic area. These regions of Vc representing parts of the body
where the patient experienced pain (and possibly dysesthesias) showed increased
bursting activity. Bursting activity was one- to three-fold greater for cells in the BAA
without RF than for control cells (i.e., those representing body parts distant from
the representation of the anesthetic part of the body). In control Vc, STT cells
fired regularly at a rate of approximately 10 spikes/s and few spike trains exhibited
high-frequency bursting. In contrast, cells recorded in BAAs showed a significantly
higher likelihood of a bursting pattern. Here, bursts were preceded by a period
of inhibition, with the initial interspike interval being less than 6 ms in duration,
becoming longer throughout the burst (i.e., decreasing number of action potentials
in the burst), a pattern typical of bursts associated with Ca2 spikes (as seen in sleep)
and involving a low-threshold rapidly inactivating Ca2 current. Moreover, cells
in the BAA region without RFs had longer preburst intervals (i.e., longer periods
of silence before a burst) and lower primary event rates (i.e., action potentials
outside bursts). In view of their inverse correlation, these cells were believed to
have tonically decreased firing rates between bursts. The most intense bursting
was found in cells that appeared to be located in the posterior aspect of the Vc
core and in the posteroinferior area (Lenz et al. 1994), where nociceptive STT
terminations are most dense (Lenz and Dougherty 1997). Thermal pain-responsive

Pathophysiology: Human Data

cells appear to be more frequent posteroinferiorly to Vc core, with warmth and

cold coded cells contiguous, but separate (see references in Hua et al. 2000).
The increase in spontaneous thalamic activity was more pronounced with more
complete interruptions of somatosensory input from a particular body part. In
further microstimulation studies (Lenz et al. 1998) of 12 neurogenic pain patients
(CPSP n 4, SCI CP n 4; Lenz et al. 1994, and PNP [n 4]; controls:
10 movement disorder cases) in parts of the thalamus representing the painful area
(both the core and posteroinferior areas of Vc), there was an increase in the number
of sites where pain was evoked by stimulation, with a corresponding decrease in the
number of sites where non-painful thermal (warm and cold) sensations were
evoked. Yet, the percentage of sites where pain or thermal sensations were evoked
was not significantly different between parts of thalamus representing the painful
and non-painful parts of the body (2%). Thus, despite the central body image being
relatively constant in the face of altered input, a reorganization occurs so that cold
modalities are relabeled to signal pain in the thalamus of patients with CP, possibly
explaining cold hyperalgesia; spontaneous bursting activity at these sites may be more
likely to produce the sensation of pain. In CP patients too, the number of sites where
cold was evoked was significantly lower than in controls, whereas the number of sites
where warmth was evoked was not different from controls (Lenz et al. 1994);
moreover, there was a significant increase in the number of sites where pain was
evoked, but no significant difference from controls in the number of pain sites plus
thermal sites.
Findings in mixed series
1. Pain and burning can be elicited in CP/PNP (but not non-pain) patients
by stimulating the STT in Vc (Hassler and Riechert 1959; Levin 1966), the
mesencephalon (Nashold et al. 1974; Sano 1977; Tasker et al. 1983), thalamic
radiations (Albe-Fessard 1973; Koszweski et al. 2003) and SI (Hamby 1961; Dierssen
et al. 1969). In this latter case, the response is obtained only in an area related to
a deafferented portion of the body (while the same stimulation in an area related to
non-deafferented body parts gives only the usual paresthesias), mimicking the
patients spontaneous pain (in the same body part as their own pain).

2. Epileptifom discharges related to pain paroxysms have been recorded in the lateral
mesencephalic tegmentum inferior and posterior to the intralaminar nuclei in
patients with PNP and CP, possibly at the site of termination of the spinomesencephalic tract (Iacono and Nashold 1982).
3. Toth and collegues (1984) examined neurogenic pain (including 3 thalamic CP
cases) and non-pain patients. They studied Vc, CM, pulvinar and mesencephalic
reticular formation, with stereotactically positioned electrodes. Unlike non-pain
patients, in patients with CP, the spontaneous activity in Vc and CM was strikingly
dysrhythmic, contained many sharp steep waves and the amplitude was pronounced,
sometimes more than in the cortical activity. The activity contained bursts composed
of sudden spike-like waves. By stimulating Vc or CM with single stimuli, in the
others, 46 Hz waxingwaning steep potential series could be recorded. During
100 Hz/500 ms train stimulation in Vc and CM, typical electroconvulsive paroxysmal

259

260

Central Pain Syndrome

activity occurred which was strictly localized within these structures. Only slight
traces appeared in the frontoparietal cortical activity (unlike Guecer et al. 1978).
These changes were most pronounced in phantom pain (4 patients), but could also
be observed in CP. In CP, the spontaneous and evoked electrical activity in the
specific and non-specific thalamic nuclei was characteristically paroxysmal and could
be strongly enhanced from each other (Vc-medial thalamus autokindling).
4. Taskers group in Toronto published an impressive series of papers on the topic.
These authors (Hirayama et al. 1989) performed single-unit analysis of spontaneous
neuronal activity in 3 patients with thalamic CP and 2 with complete cord
transection at C3 and T4, respectively (plus 4 PNP cases and 4 non-pain controls:
3 MS cases and one patient with dystonia following a supratentorial thrombotic
stroke which produced a painless DjereineRoussy syndrome). They recorded three
kinds of cells firing in bursts (types AC) and one kind not firing in bursts. (1) In
pain patients, 47% of the studied bursting cells were of type A, 42% of type B and
11% of type C. Some 43% of the cells were located in Vc, 32% in Vim, 19% in Vcpc,
4% in Vop and 2% in zona incerta. A total of 22% of bursting cells had cutaneous
RFs. In other words, bursting cells typically fired at interspike intervals of 12 ms and
interburst intervals of 50 ms. Microstimulation at sites where bursting cells were
recorded usually induced no response. Bursting cells tended to be located in Vc and
Vcpc (sites in pain patients believed to be in Vim could actually have been in Vc).
(2) In non-pain patients, 59% of bursting cells were of type A, 23% of type B, 18% of
type C. Fifty-three percent of the cells were located within Vim, 35% in centrolateralis
intermedius, 6% in Vc and 6% in Vop. None had cutaneous RFs or responded to
movements. Thus, bursting cells were rarely encountered in Vc, and those bursting cells
encountered elsewhere tended to have lower mean firing rates and longer interspike
and interburst intervals. Stimulation in Vc never induced pain. Although it was
concluded that the Vc region of pain patients (CP and PNP) contained many more
bursting cells than the comparable region in non-pain patients, with different
characteristics than bursting cells in non-pain patients, It is not possible to
determine whether the bursting cells recorded in pain patients have anything to do
with the pain the patient experiences.
They (Gorecki et al. 1989) reported thalamic exploration in 39 patients:
13 thalamic CP cases, 10 SCI pain cases, 4 postcordotomy pain cases and 11 PNP
cases. Macrostimulation was carried out in the first 23 cases, with microelectrode
recording and microstimulation performed in the last 16 cases. In these latter cases,
abnormal neuronal firing was recorded in all, as spontaneous bursts of action
potentials. The interburst interval was of the order of 50 ms; 76% of bursting units
did not have RFs. Stimulation at 8% of the sites where bursting units were recorded
induced burning or pain, being found both in close proximity to or remote from
units subserving deafferented dermatomes. The time course of appearance of these
units could not be determined. Non-pain patients also demonstrated bursting cells
with intervals of the order of 200 ms, burst frequency of approximately 5 Hz, usually
located more anterior and dorsally with respect to Vc. Unlike normal patients, in
17 cases, 16 of whom had a clear history of hyperpathia or allodynia, stimulation
in Vc elicited painful sensations, often reproducing the patients particular pain

Pathophysiology: Human Data

syndrome. In 12 cases, neuronal recordings at the stimulation site indicated that

the neurons had low-threshold mechanoreceptive fields corresponding to the
pain location and to the dermatomes affected by sensory changes, a response most
frequently obtained in Vc. The induction of pain was thus more frequent in patients
with allodynia and/or hyperpathia.
Altered thalamic somatotopy was observed. They divided the different thalamic
maps into four categories: normal, empty (when there was a general lack of response
to stimulation or lack of RFs over a large number of trajectories or when there
were only lemniscal or spinothalamic tract responses in locations at which units
with receptive fields would be expected), displaced (thalamic units possibly shifted
by atrophy or sprouting at the sites of a lesion or by altered ventricular size) and with
abnormal receptive fields. The majority of patients with thalamic CP (8/13) had an
empty thalamus. At least one patient with a thalamic infarct, but no CP, demonstrated a typical empty thalamus. In two patients, the somatotopic organization was
found to have a relatively normal sequence, but individual responses were located in
sagittal planes more lateral than expected. In 5 cases (2 CCP and 1 thalamic CP,
2 PNP), somatotopic mapping demonstrated abnormal receptive fields. One patient
with C5 clinically complete spinal cord transection had extensive RFs over the
occiput and the back of the shoulders (a location where RFs have rarely been found),
corresponding to the border of the deafferented region; in particular, the representation of the external ear, occiput and neck occupied 1.5 mm of a trajectory through
the part of the thalamus where the hand, anesthetic as a result of the spinal injury,
would normally be represented, versus a 0.10.2 mm trajectory length in movement
disorder cases. In this patient, there were also statistically significant differences
in neuronal firing patterns in the deafferented region of the thalamus, compared
with the presumably normal region of the thalamus (patient included in Lenzs series
discussed above). Two patients had wide areas of bilateral as well as ipsilateral
representation with bilateral pain induction on stimulation. The remainder of the
patients had normal maps with a propensity for SCI patients to be in this category
(6/10). These three types of altered thalamic somatotopy were present in patients both
with and without pain states.
They (Rinaldi et al. 1991) observed bursting in PNP and CP (2 cases), occurring
in two patterns, short bursts of 26 spikes every 14 s or a long burst of 3080 spikes,
at an average rate of a burst every 14 s. This activity was found concentrated to the
lateral aspect of MD, CL and only a small part of CM-Pf complex.
In an excellent study, this group (Parrent et al. 1992) reported on two patients
with massive suprathalamic infarcts. Their first case, a 58-year-old woman, suffered
a right hemispheric infarct following carotid endarterectomy. Shortly thereafter,
she developed left hemibody CP. A cordotomy was ineffective. The pain was
constant, burning, particularly significant in the shoulder. Aside from motor deficits,
there was marked sensory loss on the left side, with preserved, though reduced,
vibration sense in the left hand. There was no hyperpathia, bar a suggestion of cold
allodynia in the left shoulder area. MRI showed parenchymal loss in the distribution
of the right sylvian artery, with T1-hypointense areas in the right periventricular
region. The right cerebral peduncle and thalamus were atrophic. Stereotactic exploration of the right thalamus with the patient awake and unsedated and exhaustive

261

262

Central Pain Syndrome

microrecording plus micro- and macrostimulation of Vc and medial thalamic

nuclei revealed no motor or sensory responses of any kind and no receptive fields
were recorded. PVG stimulation produced no subjective sensations or effect on the
patients pain and allodynia. Their second case, a 57-year-old man, suffered a right
hemispheric infarct. Almost immediately following the stroke he developed CP.
Constant sharp pain was experienced in the left shoulder and hand and in the lower
back and left hip (worse in the latter two), with spontaneous exacerbations occurring
every two minutes; steady burning pain affected the medial left thigh, knee and
foot and cramping pain the left thigh and calf. Aside from motor and other deficits,
there was a diminished to absent appreciation of light touch, pinprick and vibration
in the entire left side of the body. There was allodynia to light touch and cold stimuli
on the entire left side, and hyperpathia of left limbs and face. CT showed a massive
infarct in the right sylvian artery distribution. Stereotactic exploration of the right
thalamus with the patient awake and unsedated and microrecording plus micro- and
macrostimulation obtained no motor or sensory responses. No stimulation-evoked
responses were obtained in the right PVG region. Exploration of the left PVG
obtained the typical stimulation responses of this region as well as acute relief of
the patients allodynia and hyperpathia. They concluded for a major role of the
thalamus ipsilateral to pain.
Tasker and colleagues (1994) observed bursting cells in 64% and somatotopic
reorganization in all of 29 CPSP (thalamic, suprathalamic and brainstem) patients.
Recordings showed a lesion could leave deafferented structures in neutral, but
capable of electrical and (therefore presumably) intrinsic stimulation to possibly
produce pain. Macrostimulation of the tegmental reticulothalamic pathways
(and medial thalamic nuclei), normally unresponsive to stimulation, at threshold
effective for ML/STT stimulation, induced a widespread nonsomatotopographically
organized burning or pain sensation (mimicking the original pain) extending beyond
the involved dermatomes, often similar to that from which the patient suffered
(5 brainstem CPSP, 1 MS, 1 CCP). Stimulation tended to be painful in patients
with evoked pain (14/16) but not without (1/4), even in the absence of contralateral
functional SI (or massive hemispherectomy-like lesions); the reticular system was
thus implicated in allodynia, ipsilateral structures in the mediation of constant pain
(Tasker et al. 1983; Tasker 2001a).
Thalamic reorganization following denervation was tested by studying thalamic
somatotopy (microrecording/stimulation) in 61 patients: 5 groups were compared
according to body part in patients with pain in the deafferented body part and in
controls (movement disorders). PNP and CP were considered together (Kiss et al.
1994). Trunk representation (RF) was significantly larger in patients with legfoot
deafferentation than in those without; however, microstimulation induced paresthesias in the face from a significantly larger thalamic area in facially denervated cases
than controls (i.e., face RFs increased, but maintained small discrete PFs not extending into other body parts). There were no significant differences in the representation
of the other body parts in the 5 groups. In the leg-deafferented-only group, the
deafferented cells responded to afferent input from an adjacent body part, yet
retained their original connections to the cortical representation of the deafferented
body part. In face-deafferented patients, deafferented cells ceased to respond to

Pathophysiology: Human Data

peripheral inputs, yet maintained their thalamocortical projections to the original

body part representation. In some patients, deafferented cells could both stop
responding to peripheral input and communicate meaningfully with their cortical
target.
The Vc core (but not other nuclei more ventroposterior to Vc) was studied
in 5 thalamic, 3 suprathalamic, 2 internal capsule and 3 cortical CP cases (versus
23 non-stroke pain and 24 movement disorder patients) with stereotactic microrecordings (Davis et al. 1996). Microstimulation in the tactile core of Vc commonly
evoked paresthesias, while threshold stimulation never or rarely (2%) evoked pain in
non-stroke and movement disorders patients, respectively. By contrast, in CP,
28% of Vc sites microstimulated evoked painful sensations at threshold (suprathreshold stimuli did so at 46% of Vc sites in CP versus 8% in other pains and 12% of
movement disorders cases). There was no significant difference between the paresthesia thresholds of non-CP patients and motor patients, but these were elevated
two-fold in CP patients, except 4 (2 patients with particularly small thalamic lesions
and 2 patients with small cortical lesions). However, stimulation thresholds to
elicit pain were similar in all patient groups. CP patients most often noted the
stimulation-evoked pain as a nondescript pain (33% of sites) or painful burning
sensations (43% of sites), shocking (10%) or sharp (14%). In control groups, pain
was elicited only with stimuli suprathreshold for paresthesias. Most common
with suprathreshold stimuli was an unpleasant (or sometimes shocking) feeling in
the non-CP pain group (61% of sites) and movement disorders (45%). The burning
sensation so often reported by CP was never reported by the movement disorder
patients and at only two sites in the non-CP patients. Interestingly, qualities of evoked
pain in pain patients did not necessarily relate to the quality of the patients ongoing
chronic pain. Pain could be evoked at sites throughout tactile Vc, although most sites
were located in the ventral two-thirds. Microstimulation within Vc almost always
evoked a response, even in the presence of suprathalamic infarcts (and also with
thalamic lesions). Vc stimulation in 62% of CP patients evoked pain: this was
not related to allodynia, since pain was evoked in patients with (4/7) and without
(3/6) it. In some CP patients, pain was evoked throughout the electrode trajectory
within Vc, a clustering not seen in the other two groups. At some Vc sites in CP
patients, stimulation up to maximum current (up to 100 mA) did not evoke any
sensation. Suprathreshold stimuli in CP converted only a few responses from
paresthesia to pain. In some patients with pain, there appeared to be a decrease in cell
density in regions representing body parts whose afferents had been damaged.
Although RF/PF mismatches in non-pain patients were noted for nearly half of Vc,
they were minor or simple size discrepancies; stimulation at only 9% of Vc in these
control patients resulted in gross mismatches. The total number of RF/PF mismatches was significantly greater in both pain groups compared with motor group,
due to a greater increase in gross rather than minor or size mismatches in the pain
patients. The proportion of all mismatches was the same in the non-CP and CP groups
and size mismatches were similar between CP and non-CP patients.
In a major study, Radhakrishnan and colleagues (1999) compared the incidence of
bursting in Vc of patients with neurogenic pain (including CPSP and SCI, whose
numbers were not specified) and motor disorders. The burst indices (i.e., the number

263

264

Central Pain Syndrome

of bursting cells per track) in the pain and non-pain groups were not significantly
different from each other. Low-threshold Ca2 spike-evoked bursts (with shortening
of the first interspike interval, an increase in the number of interspike intervals in the
burst and progressive prolongation of successive interspike intervals) were identified
in 57% of bursting cells in pain patients and 47% of non-pain patients, suggesting
no definite rapport with pain. Only a few cells of the bursting kind were located
in Vc, the majority being anterodorsally and ventroposteriorly to it (see also Ohye
and Narabayashi 1972).
Finally, they (Manduch et al. 1999) did microelectrode recordings in 40 movement
disorder and 37 chronic pain patients through Vc and regions ventroposterior
to it. Stimulation evoked painful or innocuous thermal sensations at 2.9 and 4.7%,
respectively (5023 stimulation sites). A total of 77% were located ventroposterior to
Vc and of these 74% were located in or medial to the face/hand representation border
in Vc. No significant differences were noted between controls and non-CPSP cases
in the incidence of pain and temperature sites. Instead, the incidence of pain sites
was higher in CPSP cases (n 11) compared to the other 2 groups (9.5% versus
2.5% in the ventroposterior region of Vc and 15.1% versus 1.4% in Vc). In contrast,
the incidence of thermal sites was lower below Vc in CPSP than in the other 2 groups,
but not different in Vc.
5. Yamashiro and colleagues (1991) made microrecordings in the Vc of 2 patients
with SCI, 1 with CPSP, 1 with MS-associated CP and 4 PNP cases. Epileptiform
discharges from hyperactive neurons were recorded and two firing patterns seen. One
showed regular firing which had 35 trains of epileptiform grouped discharges with
a frequency of 45 Hz. The latter showed continuous firing. These hyperactive
neurons were distributed in Vc, Vim and Vop and may have received facilitation
from SI/MI.
6. Jeanmonod and colleagues (1996) recorded unit activities from the thalami
of 74 patients with CP and PNP. Some 99.8% of their medial thalamic units did not
respond to somatosensory stimulation (in contrast to a few other studies; see Lenz
and Dougherty 1997). In addition to their unresponsiveness, half of the units showed
a striking bursting (45.1%) activity (rhythmic: 25%; random: 30%) not due to sleep,
as all patients were fully awake during surgery. The rhythmic-random low-threshold
Ca2 spike (LTS) bursting units were considered abnormal and were found
distributed throughout the posterior half of CL. The rest of their sampled units
displayed unresponsive sporadic activities. Many of them exhibited occasional LTS
bursts. LTS bursts displayed a theta rhythmicity, with a mean interburst discharge
rate of about 4 Hz. In patients with intermittent pain without steady component,
they made recordings only during pain-free periods, and never showed a large
amount of LTS bursts, as can be the case in patients with steady pain.

interpretation: Anomalous activity at several CNS levels is observed in CP patients.

However, most anomalies are seen both in PNP and CP patients, making them
unspecific; most importantly, they are not invariably found. Some findings involve the
thalamus in the genesis of CP: (1) an increased incidence of pain evoked at threshold

Pathophysiology: Human Data

in Vc (core and shell) in CP versus PNP or other controls (see also Hassler and
Riechert 1959; Levin 1966; Mazars et al. 1974); (2) a likely role of thalami ipsilateral
to CP; (3) thalamic involvement in cold allodynia (see below). Different participation
of Vc, Vim and CL may justify different qualities of CP. Interestingly, Lenz and
Dougherty (1997) reported that sensations are more likely to be referred to deep
structures at stimulation sites in Vc posteroinferior areas more than in the core.
Along these lines, useful information should accrue by studying central pruritus
patients: itch is a purely cutaneous sensation and might elucidate central mechanisms
of cutaneous anomalous sensations (Canavero et al. 1997).
The reticular formation (and related propriospinal cells and fibers in the DREZ) is
also likely involved (see Chapter 8). Somatotopic rearrangements (such as expansion
of adjacent regions into denervated) and burst firing (Box 7.3) seem to be the result
of denervation injury, and not a correlate of pain (unlike, possibly, phantom pain),
since they can be observed in non-pain conditions (Jeanmonod et al. 1996; Tasker
2001b; see Chapter 8). Since Vc stimulation evokes tactile allodynia more commonly
in CP than non-CP pains (Davis et al. 1996; Lenz et al. 1998), pain more frequently
in those with hyperalgesia than in those without and in the representation of the part
of the body where the patient experiences hyperalgesia than in the representation
of other body parts (Lenz et al. 1998), the findings discussed may have a special
relevance to allodynia (see Chapter 8).

2. Evoked potentials studies

1. Mauguie`re and Desmedt (1988) differentiated four types of CP of thalamic origin
by somatosensory evoked potentials (SEPs), which explore dorsal columnmedial
lemniscal (DC/ML) function: group 1 had no CP, but complete hemianesthesia and
loss of cortical SEPs on the affected side (analgic thalamic syndrome); group 2 had
CP, severe hypoesthesia and loss of cortical SEPs; group 3 had CP and hypoesthesia,
with cortical SEPs present, although reduced or delayed on the affected side; group 4
had CP with preserved touch and joint sensations and normal SEPs (pure algetic
thalamic syndrome). All their 30 patients presented a thalamic lesion on CT. SEPs
did not tell apart groups 1 and 2, but separated these two groups from group 3, in
whom cortical SEPs were present.
2. Wessel and colleagues (1994) studied 18 patients with a single ischemic thalamic
lesion, who had somatosensory disturbances and/or CP in the opposite hemibody, by
correlating their clinical symptoms, SEPs and CT imaging findings. Patients were
divided into three groups: (1) those with somatosensory deficits, CP, and abnormal
SEPs, which comprised two thirds of the patients (classic thalamic pain syndrome);
(2) those with somatosensory deficits, no CP and abnormal SEPs (analgetic thalamic
syndrome), with a 1-year follow-up; and (3) those with almost normal sense
perception, CP and normal SEPs (pure algetic thalamic syndrome). Six of the 8
patients with the analgetic syndrome had a posterolateral thalamic stroke in the
territory of the geniculothalamic artery, which includes Vc, whereas groups 1 and 3
had CT evidence of paramedian or anterolateral thalamic lesions.

265

266

Central Pain Syndrome

Box 7.3 Does bursting signal CP?
All thalamic relay (e.g., Vc) cells respond to excitatory inputs in one of two different modes: burst and
tonic (Sherman and Guillery 2004). In burst mode, the inward T-type (IT) Ca2 channel in soma and
dendrites is activated and an inflow of Ca2 produces a low-threshold spike (LTS) that in turn usually
activates a burst (usually less than 25 ms long) of conventional action potentials. After about 100 ms
or more of depolarization, the IT inactivates and the cell fires in tonic mode; after about 100 ms or more
of relative hyperpolarization, inactivation of IT is alleviated and the cell fires in burst mode. Just
like tonic, burst firing is an important relay mode during waking behavior and could play an important
role in attention. Unlike single action potentials, which are very often filtered out, bursts particularly
coincident bursts are reliably signaled, because transmitter release is strongly facilitated. In fact,
single spikes are spontaneously emitted by neurons, creating noise (i.e., disinformation) (Lisman
1997). Bursts are particularly effective for synaptic communication in the cortex. Specifically, bursting
can record significant, but possibly minor changes in specific afferent activity (initial stimulus detection)
and use this to focus the tonic mode upon the causes of these changes for more accurate analysis.
Also, rhythmic bursts may signal no transmission, while arrhythmic bursts may indicate sensory transmission. Switching between tonic and burst firing occurs irregularly every several hundred milliseconds
to every several seconds, presumably reflecting slow changes in membrane potential that switch
IT between inactivated and deinactivated. Only 510% of synapses on Vc TC cells come from the
periphery: 30% are from local GABAergic neurons, 30% from cholinergic sources and 30% from SI
layer 6. Thus the vast majority of inputs are modulating, controlling the state of IT and thus the response
mode between burst and tonic (Sherman and Guillery 2004).
Despite several authors highlighting the importance of Ca2-related bursting activity (inside the
more general phenomenon of sensitization) in the genesis of CP, several pieces of evidence nix
this concept:
(1) Ca2 LTS bursts in the thalamus (particularly Vc and CL) (thalamic dysrhythmia) and supposed
neurophysiological correlates thereof (i.e. theta/beta bands of juxtaposed cortical activity) are not
specific to CP, and have been consistently observed in the same nuclei in PNP patients, as well as
CNS disorders without a pain component (Jeanmonod et al. 1996; Llinas et al. 1999: no CP cases
were studied in this latter paper). Also, loss of corticothalamic input (Llinas et al. 1999) believed
by some authors to produce CP-associated electrophysiological anomalies, actually abolishes CP
(see section on reports of sudden disappearance of CP).
(2) Although it seems more prominent in neurons with representation areas in the anesthetic part of
the body, Ca2-related bursting is also found in normal awake controls, during slow-wave sleep
and also during anesthesia. The proportion of intrinsically bursting cells in the intact cortex is about
1520% (Steriade 1999; Sanchez-Vives and McCormick 2000) and many such bursting cells are
found in the thalamus (Tasker 2001b). Most importantly, there is no coincidence between pain
sensation in CP patients and moment of burst discharge (Ohye 1998) and following anterolateral
cordotomy CP is not usually felt in areas of the body as surmised from, for example, Lenzs
speculations (Beric et al. 1988).
(3) Bursting, as CP sensation-related activity, due to STT injury, cannot explain immediate-onset CP (is
bursting immediate?) and also cases in which there is no clinically evident STT-mediated sensory
loss (e.g., Stoodley et al. 1995).
(4) The balance of excitatory and inhibitory inputs leading to Ca2 spike associated bursting is unclear
(overinhibition? Loss of excitation?) and may depend on the presence or absence of RFs.
(5) Lenzs view that bursting and absence of RFs in the Vc BAA of SCI CP patients is due to decreased
tonic NMDA excitatory drive with attendant hyperpolarization collapses on STT input being nonNMDA mediated. Also, the fact that Ca2 bursting may be decreased by norepinephrine and
increased by acetylcholine (i.e., amitriptylines profile) is meaningless in view of poor efficacy
in CCP (Chapter 5).

Pathophysiology: Human Data

(6) In patients with spinal transection, the painful area overlaps with the area of sensory loss (Lenz
et al. 1994), making bursting the result of sensory loss rather than pain.
(7) A 1 mm increment of electrode insertion in an area of spontaneous discharge can result in
an artefactual temporary increase in activation of the existing discharge patterns (Andy 1983).
Models explaining bursting in relation to CP collapse for all these reasons (e.g., Jeanmonod et al.
1996). Moreover, those models, even in these authors minds, have much more difficulty in explaining
CP than PNP. The anatomical background too is unsupported: bidirectional TRN interconnections
between pain-related CL and Vc cells, with back-and-fro exchange of waves of inhibition, starting
from the less sensory-input-deprived CL exciting TRN, is presently unsubstantiated (Steriade et al.
1997).
Bursting can be a normal condition of thalamic functioning in awake humans. Possible increments of
bursting in certain locations can be explained away as an injury-related disorder of normal thalamic
oscillatory mechanisms (Ohye 1998; Tasker 2001b). On the other hand, Kim and colleagues (2004)
proposed that it may be part of a robust pain-relieving mechanism. T-type Ca2 channel activation in
Vc can activate TRN cells, with subsequent hyperpolarization and rebound burst spikes again in TC cells
through reciprocal VcTRNVc connections; hyperpolarization and/or burst sequences can contribute to
sensory inhibition by reducing the responsiveness of TC neurons. Specifically, because a burst has
a long refractory period (170200 ms), bursting sequences might actually prevent rapidly recurring
sensory signal inputs to TC relay cells. Inactivation of this Ca2 channel and thus bursting interferes
with sensory gating of pain. The nociceptive dampening/filtering role of the thalamus had
been hypothesized by several past authors (e.g., selective filter [Lhermitte 1933]; thalamic
function . . . with . . . an inhibitory effect of normal afferent impulses [Botterell et al. 1954]).

I N T ER P R E TA T I O N :

Complete interruption of lemniscal transmission through Vc up to

parietal cortex does not necessarily release the mechanisms underlying CP, refuting
past theories of deficient lemniscal inhibition of nociceptive STT conduction. Most
importantly, both series agree that complete destruction of Vc and possibly other
nuclei may be incompatible with the occurrence of CP. Ohye (1998) reached the
same conclusion. Thus, the sensory thalamus is necessary for CP to arise.

RESULTS OF NEUROABLATION
Current ablative techniques have no or only a limited role in the management
of CP. On the other hand, they provide invaluable insight into the mechanisms
subserving CP (see Table 7.1).
Interpretation

a.

PRE- AND POST-CENTRAL GYRECTOMY (FIRST PROPOSED BY LERICHE 1937):

Limited
cortectomies relieved some cases for years, although others were failures. In the CP
case reported by Lende and colleagues, cortical removal extended up to the border
of the motor and sensory representation of the hand area and down to the sylvian
fissure, with excision of the operculi of the pre- and post-central gyri, and exposing
the insula. Thus, effective cortectomies should likely include not only SI, but also
SII/insula and even MI. SI-MI coactivation in metabolic studies underlies the
concerted effectuation of interrelated sensorimotor functions (see also Penfield and
Jasper 1954; Libet 1973). At least some failures can be explained away by the wide
variability in somatotopy in individuals and somatotopic differences not only

267

268

Thalamic lesion (1 pt)

CPSP (1 pt)

Thalamic lesion (1 pt)

CP, brain injury (1 pt)

CP, thalamic (2 pts)

CP (1 pt)

CPSP (1 pt)

CPSP (1 pt)

Stone (1950)

Penfield and Welch (1951)

Lewin and Phillips (1952)

Erickson and colleagues (1952)

Spiegel et al. (1952)

White and Sweet (1955)

Biemond (1956)

Relief for 18 mos, then relapse

Relief for months until relapse

Limited (2 cm) SI cortectomy

insulectomy

No relief

Relief for 2 years in both

SI gyrectomy

SI gyrectomy

SI in toto gyrectomy

Relief for 4 years

Relief, then relapse

MI (atrophied) gyrectomy
Excision of the cerebrodural scar
underlying subcortical cyst

Relief for 18 mos, then relapse

Relief for at least 14 mos

Relief for 2 mos

0%

Relief at 14 mos, except arm/

hand pain relapsed after 5 mos
Relief until death mos later

SI gyrectomy

Subpial section of the postcentral gyrus

Procaine injection into SI

SI gyrectomy

Leriche (1949)

SI gyrectomy

SI gyrectomy

0%

0%

Corpus-callosectomy of parietal
associative fibers
Tumor excision

SCI (bony spur at C6)

CP, glioma of the left

hemisphere

Horrax (1946)

0%

Efficacy/(follow-up)

Cortectomy SI large parts of

superior and inferior parietal gyri

Procedure

CP, rolandoparietal glioma

Thalamic lesion (juxtainsular

lesion affecting the corona
radiata)

Cause of central pain

Dimitri and Balado (quoted by

David et al. 1947)

Pre- and postcentral gyrectomies

Author(s)

TABLE 7.1. Results of neuroablative procedures

No benefit from previous cervical

cordotomy
At autopsy: softening in the parietal
and insular cortex, degenerated fiber

Convulsive seizures preceded by an

aura including torturing, deep,
gnawing pain in the wrist and hand,
spreading to the left limbs and left
side of the face

SI stimulation triggered patients

pain

No benefit from previous cervical

cordotomy

At autopsy, iuxtainsular lesion in

corona radiata

Notes

269

CP, postcordotomy

CPSP, brainstem (1 pt)

White and Sweet (1969)

Lende et al. (1971)

Thalamic syndrome (2 pts)

CCP (1 pt)

CP, thalamic (1 pt)

CPSP (1 pt)

CP, brain (1 pt)

Mesencephalotomy-induced CP
(1 pt)

CP, brain (1 pt)

CP (5 pts)

Guillaume et al. (1949)

Wertheimer and Mansuy (1949)

Freeman and Watts (1950)

Scarff (1950)

Gaches (1952)

Drake and McKenzie (1953)

Petit-Dutaillis et al. (1953)

Le Beau et al. (1954)

Psychosurgery

Pure cortical CP (1 pt)

Hamby (1961)

Bilateral BA 910 topectomy

Frontal lobotomy

Frontal lobotomy

Frontal lobotomy

Left prefrontal lobotomy

Prefrontal lobotomy

Frontal lobotomy

Frontal lobotomy

Cortectomy of SISII and MI

Transpial incision five mm

deeper than the gutters of the
gyri along the posterior edge of
SI and over three contiguous
parietal gyri. Removal of the
cortex and adjacent U-fiber
areas of the white matter
SI gyrectomy (bar face sector)
down to the sulcus cinguli

Almost complete, but pain

admitted on interrogation/
follow-up 4 years;
0%/follow-up 6 mos

Not available for review

No

Improved, but not abolished

Good relief, relapse at 4 mos

Relief

Indifference toward pain,

which was still present and
severe (one resumed some
activity after surgery)
0%

Relief for 20 mos

0%

Relief for 10 years

(continued)

In other pains, unilateral lobotomy

may relieve bilateral pains

Pain not relieved by previous

complete trigeminal rhizotomy

Pain evoked by SI stimulation

bundle tracing to thalamus (VPM)

through the internal capsule, cell
loss in VPM
Painful, prickling sensations in the
arm and hand elicited from
stimulation of SI

270

Foltz and White (1966)

CP, brain (1 pt)

Rostral cingulumotomy

Bilateral prefrontal lobotomy

Frontal operation

Unilateral frontal leukotomy

CCP (1 pt)

Tabes dorsalis (2 pts)

CP (1 pt)
SCI CP (3 pts)

Fractionated radiofrequency
frontomedial leukotomy

CCP, postcordotomy (4 pts)

Wycis and Spiegel (1962)

Unilateral frontal leukotomy

CPSP (1 pt)

Constans (1960)

Bilateral orbital gyrectomy (BA

1112)

CP (2 pts)

White and Sweet (1955, 1969)

Prefrontal lobotomy

1 fair at 4 yrs and 1 poor at 3

yrs (unilateral); 1 excellent at
1 yr, then fair at 2.5 yrs
(bilateral cingulumotomy)

0%

Gratifying (follow-up: not

available)
Failure, then success at
2nd operation
0%
Pain sometimes felt, but not
bothering; total disappearance
over 2 yrs until death another
2 years later (patient had
neglect)
Burning relieved, but PNP-associated hyperpathia 0%
100% immediate relief;
total relapse at 2.5 mos
0% (2 pts)
Not complete relief, but no
longer in need of analgesics for
16 years
Unsatisfactory

0% over 2 weeks

Unilateral frontal lobotomy

Bilateral orbital gyrectomy

Complete relief after 2nd

surgery/follow-up not specified
1020% relief for 2.5 yrs

Efficacy/(follow-up)

Unilateral BA910 topectomy

Procedure

CP, SCI (not available)

Cause of central pain

Botterell et al. (1954)

Le Beau et al. (continued)

Psychosurgery (continued)

Author(s)

TABLE 7.1 (continued)

Transitory relief with mesencephalotomy, then relapse

Notes

271

Tabes dorsalis (2 pts)

CP, brain (2 pts)

CP, SCI (1 pt)

CP, brain (3 pts)

CP, SCI (1 pt)

CP, brain (1 pt)

Nashold and Wilson (1970)

Turnbull (1972)

Bouchard et al. (1977)

Jefferson (1983)

Ballantine and Giriunas (1988)

Tasker (1990)

Pillay and Hassenbusch (1992)

Bilateral stereotactic
cingulotomy

Bilateral stereotactic
cingulotomy
Bilateral stereotactic
cingulotomy
No (VAS from 9 to 8); quality of
life unchanged

No relief

No substantial relief

Reasonable relief

Benefit

Bilateral stereotactic
cingulotomy

No benefit

Contralateral cingulotomy

Relief

0%

0%

Gratifying

Ipsilateral cingulotomy

Bilateral cingulotomy

Unilateral left frontal lobotomy

Bilateral anterior capsulotomy

Prefrontal lobotomy

Followed by unsuccessful bilateral

medial thalamotomy and mesencephalic tractotomy

Previous unsuccessful cordotomy

Stereotactic chemical
hypophysectomy

CP (7 pts)

CCP (1 pt) plus another CP?

CPSP, thalamic (17 pts)

Miles (1998)

Hayashi et al. (2005)

Pituitary gamma knife radiosurgery at the border between the

pituitary stalk and gland (max.
dose 140160 Gy, 180 in 1 pt)

Hypophyseal stimulation

Posterior hypothalamotomy
(medial; III ventricle gray matter)

CP, brain and cord (at least 2 or

more BCP pts)

Mayanagi and Sano (1988,

1998); plus Amano et al. 1976
(in Japanese)
Levin (1988)

Hypophysectomyhypothalamotomy

13/17 had pain reduction

(within 48 hours): 76.5%
At longer term (41 yr):
5/13 effective relief (in 1 pt
480% relief); 4/13 fully
relapsed within 3 mos; 4/13
still relieved at 6 mos

More than 50% relief in 6, 2 still

relieved 2 yrs later; at least
2 relapses within a few mos
0%

No (025% in one CPSP at max.

follow-up 17 mos)

(continued)

Single 8 mm isocenter, 50% isodose

line covered the border between the
pituitary gland and lower part
of pituitary stalk
No complications
Numbness not improved
Neuromodulatory effect hypothesized

Several complications

Pain increased by electrical stimulation in 2 pts

Mazars et al. (1976) cite two cases of thalamic pain submitted to frontal lobotomy by Siocet and Bartich with improvement, but not abolition

CPSP (1 pt)

1 CPSP (brainstem)

Spiegel et al. (1966)

CP, SCI

Porter et al. (1966)

272
Vc

CP, brain (2 pts)

Laspiur (1956)

CP, SCI (4 limb burning

dysesthesias) pts?
CP, brain (2 pts)

Neurogenic pains

Bettag and Yoshida (1960)

Mark et al. (1960)

Davis and Stokes (1966)

Hankinson (1962)

CP, brain (4 pts)

CP, thalamic (4 pts)

Hassler (1960)

CP, thalamic (2 pts)

CP, brain (12 pts)

Talairach et al. (1955)

CP, brain (1 pt)

CP, brain (3 pts)

Spiegel and colleagues (1952)

Obrador et al. (1957)

CP, brain (1 pt)

Baudoin and Puech (1949)

Hassler and Riechert (1959)

Vc

CP, thalamic (12 pts)

Lateral plus medial nc

CM and Vc

Vc

DM (1 pt)

Vc (3 pts)

Vc, limitans and CeM

Vc

Vc

Local novocaine injection into

Vc
Vc

Vc (radioactive gold)

2 center median DM

Talairach et al. (1949)

1 center median
1 center median Vc

Procedure

CP (4 pts)

Cause of central pain

Hecaen et al. (1949)

Thalamotomies

Author(s)

TABLE 7.1 (continued)

Immediate pain relief in 75% of

pts, decreasing to 5060% after
612 mos

Partial pain relief, but recurrence after 6 mos

Yes (1624 mos)

In all, lasting relief

Yes, lasting relief

Relief, 5 weeks

Yes (in one, 100% relief, in the

latter spectacular relief)/
follow-up?
0%

Temporary (max. 4.5 mos), in

one relapse after a few weeks
Favorable relief in 50% of pts

0%

6, 75100% reliefs; 2, 50%

reliefs; 2, 25% reliefs; 2 deaths

Yes, immediate (4 mo)

Yes, immediate and complete
(f-up: 1 yr)
Yes, immediate and for at least
4 mos

Efficacy/(follow-up)

1 suicide

Thalamic hand and clonus induced

by Vc stimulation, no effect with DM
stimulation

Notes

273

CCP (conocaudal) (1 pt)

CP, brain (unspecified)

Neurogenic pain

Burning hypesthesia and spastic

hemiplegia (3 pts)

Sugita et al. (1972)

Siegfried and Krayenbuehl

(1972)

Cooper et al. (1973)

LP-pulvinotomy

Vc, intralaminar system plus DM

CeM, Pf, intralaminar, MD

Pf (bilateral) and ant. Nc.

(unilateral)
Vc and DM (unilateral)

CCP (cervical) (1 pt)

Tabes dorsalis (1 pt)

Pf (unilateral)
Vc

CP, MS (1 pt)

Pf (unilateral)

CP, brain (1 pt)

White and Sweet (1969)

Bilateral basal thalamotomy

SCI pain (1 pt)

Pulvinar

Basal thalamotomy

CPSP (1 pt)

CPSP (6 pts) out of 17 with

cancer or noncancer pain

Medial thalamotomy

CPSP (1 pt)

Kudo et al. (1968)

Medial thalamotomy

CCP (1 pt)

Spiegel et al. (1966)

CM, DM

Neurogenic pains

Bettag (1966)

Acute pain sensation not affected

1 of 9 pts with Vcpc

thalamotomy relieved
Relief in 3. No relapse

(continued)

Not available for review

No

No effect

Poor relief

Good relief

Fair relief

Good relief

100% relief for 3 wks, then

partial relapse (superficial vs.
deep pain) with allodynia,
at 3 mos
100% relief, full relapse
at 4 mos
Whole series: 8 complete
reliefs, 6 remarkable, 3 slight
pain remaining
Poor result

Persisting pain relief only

in 6/31 pts
Pain relief only in 1/4 pts
subjected to CM lesions, with or
without DM
100% relief; full relapse 1 wk
later; 100% relief after reop.
At 1.5 yrs, pain reduced
Partial relief. Late result:
indifference to pain

274
Pulvinar

CP neurogenic pain (13 pts)

CP, brain

Siegfried (1977)

Pagni (1977)

CP, SCI

Medial nc

CP

Intralaminar nc (including CM/

Pf), sometimes extending to Vc
and DM

Follow-up: 6 mos
CP difficult to control
40% good; total relief up to
14.5 yrs
Yes, dramatic initial relief in
several. Recurrence within 1 yr
in several
Total or partial long-term relief
in 12 BCP and 3 CCP
Pagnis experience with CP:
30% relief

Basal: CM /
pulvinar

CP (thalamic: 3 pts)

Mayanagi and Bouchard

(197677)
Mundinger and Becker (1977)

Efficacy/(follow-up)

Thalamic CP: follow-up 124

months. At discharge: 100% in
3, slight residual but tolerable
pain in 6, 0% relief in 1. At
follow-up: 100% in 3, tolerable
pain in 4, tolerable pain with
drugs in 2, 0% in 1.
Other CP: at discharge: 100% in
2, slight residual but tolerable
pain in 6, tolerable with drugs in
3, some relief but intolerable in
1, 0% in 2. At follow-up: 100%
in 2, tolerable in 4, tolerable
with drugs in 3, some relief but
intolerable in 2, 0% in 2

Procedure

Thalalaminotomy (i.e., CM-Pf

and CL)

Thalamic CP (10 pts)

Other CP (14 pts)

Cause of central pain

Amano et al. (1976) (and Sano

et al. 1966)

Thalamotomies (continued)

Author(s)

TABLE 7.1 (continued)

Survey. Dysesthesia can persist

unmodified. Multiple thalamic
(CM-VPL/VPM-pulvinar) and
mesencephalic coagulations may be
necessary if lesions to a single
structure are unsuccessful. Center
median lesions very effective for
thalamic pain, with long-lasting
results. Basal thalamotomies for

Some had subtle sensory alterations

Notes

275

CT-guided pulvinarectomy

CPSP (3 pts)

SCI (1 pt)

CPSP, mainly deep muscle pain

(about 40 pts)

CM-intralaminar thalamotomy

CPSP (2 pts)

Laitinen (1988; see also 1977)

Ohye (1990, 1998)

CM thalamotomy

CPSP (2 pts)

Barcia Salorio et al. (1987)

Vim (a part)-Vcpc (deep

portion) thalamotomy (i.e.,
coagulation of the isolated
hyperactive area around the
thalamic stroke lesion)

LINAC radiosurgical Vc
thalamotomy

Unilateral/bilateral center
median

CPSP (17 pts, one of which

cheiroral)

Deep pain of compressing,

burning or sometimes squeezing
nature considerably ameliorated

Good early result

Yes, immediately (818 mos)

Burning paroxysms abolished,

background pain diminished/
follow-up 6 mos
Yes (624 mos)

Relief (1, 100%) in 56%, then

full relapse within 7 mos in all

Yes (5/6 pts)

Yes

Medial (CM), some bilateral

Niizuma et al. (1982) Includes

Niizuma et al. (1980)

Yes (2/3 pts)

Yes (5/5 pts)

Basal (including Vcpc and n.

limitans portae)

CP, brain (3 pts) Postc

ordotomy/thoracotomy
dysesthesias
(5 pts)
CP, brain (6 pts). Postcordotomy
dysesthesias (1 pt)

Hitchcock and Texteira (1981)

Yes, immediate complete in 6

pts, almost complete in 7 pts,
good in 1 pt
At 3.510 yrs: 4 pain-free,
4 almost pain-free, 3 sufficient
pain relief, 3 failures
Cases with follow-up 4 5 yrs:
1 pain-free, 2 almost pain-free,
3 sufficient pain relief, 2 failures

Pulvinar (bilateral if needed;

supranucleus pulvinaris medialis nc lesion in all cases)

CP (14 pts)

Yoshii and colleagues (1980)

(continued)

No true effect on paresthesia and

numbness

CM-intralaminar and pulvinar lesions

highly effective for CP. However, in a
mixed series of cancer and neurogenic pain, only 29% were pain-free
after 2.5 yrs

CM thalamotomies deemed superior,

particularly if bilateral, to basal
thalamotomies (better pain relief
and fewer side effects).
Very high rate of complications

brainstem lesions. Long-term results

with CM-Vc, intralaminar and DM
lesions generally unsatisfactory
No bearing on final outcome from
bilateral lesions

276

CP (thalamic) (3 pts)

Young et al. (1995)

Frighetto et al. (2004)

CPSP (thalamic and putaminal)

(2 pts)

Hirato et al. (1995)

CPSP (MCA stroke and thalamic

stroke) (2 pts)

SCI pain (1 pt)

CP, parietal cortex (5 pts),

thalamus (3 pts), brainstem
(4 pts), spinal cord (12 pts)

CPSP (9 pts)

Cause of central pain

Jeanmonod et al. (1996, 2001)

Thalamotomies (continued)

Author(s)

TABLE 7.1 (continued)

Radiosurgical CM/Pf
thalamotomy

Gamma knife medial

thalamotomy

Radiosurgical Vim thalamotomy

Medial thalamotomies
(if necessary, lesion ipsilateral
to pain)

Vim and/or CL thalamotomy

Procedure

B: some drug reduction, allodynia improved, 3 yrs later drugs

only twice a week

A: Vim thalamotomy: some

relief, relapse, radiosurgical Vim
thalamotomy, relief
B: Vim thalamotomy: poor relief,
gamma thalamotomy good relief
(not abolition) for 3 mos
Median follow-up for whole
group of 20 mixed pains pts;
about 1 yr relief seems to have
been obtained in at least some
A: immediate relief, relapse
at 4 mos (relieved by MCS)

Yes, 50100% relief in 40% of

BCP pts and 38% of SCI pts.
Relief was best for evoked and
intermittent pain and superficial
pain, poorer for steady pain
(which lingered on in more than
half the cases) and deep pain
One CCP patient referred by us:
0% relief ( complications)

Satisfactory relief in 4/9

Efficacy/(follow-up)

No 100% abolition; effect on pain

before onset of necrosis (!);
necroses 3.5  5 mm and 8.5  7
mm (too large to have exclusively
targeted CM-PF)

Relief seen in both after 36 mos

(!?)

Generally without postop. somatosensory including pain deficits; in

several, postop. improvement of
somatosensory deficits

Relief only of deep pain and not

superficial or dysesthetic pain

Notes

277

CP (5 pts?)

CPSP (25 pts)

CP, tumor (1 pt)
Postcordotomy dysesthesia-PCD
(1 pt)
Tabetic pain-TB (1 pt)

Schvarcz (1977)

Amano et al. (1980, 1986,

1992)

Rostral mesencephalic reticulotomy (highly selective lesion in

the medialmost portion of the
midbrain reticular formation,
medial to the STT which is not
lesioned unlike Nasholds
procedure) contralateral to pain
in all cases

Combined mesencephalotomythalamotomy-cingulotomy
Mesencephalotomy

Medial lesion

CP neurogenic pain (2 pts)

Tabes dorsalis (1 pts)

Lateral plus medial lesions

Turnbull (1972)

STT

CPSP (1 pt)

Orthner and Roeder (1966)

Includes Roeder and Orthner
(1961)
Gioia et al. (1967)

CPSP (thalamic) (1 pt)

(continued)

Results confirmed in 1992.

64% complete or near complete
pain relief. No postop. dysesthesias
One of the pts relieved 100% at
11 yrs noticed at year 78 tactilethermoalgesic anesthesia of left
hemisoma
Group 1: 2 complete reliefs,
3 partial reliefs at 5070 mos
Group 2: 6 complete reliefs,
9 almost complete reliefs,
6 partial reliefs at <50 mos
PCD 0% relief. TB almost
complete relief at 57 mos

Not available for review

3 mesencephalotomies plus
thalamotomies: 1 complete relief for
10 yrs, 1 partial relief, 1 transient
indifference

4 pain reliefs at 624 mos

1 modest relief

Poor

Almost complete relief for

26 mos up to death

0%

CCP (3 pts)

Helfant et al. (1965)

0% relief
1 complete relief for 1 yr,
1 transient relief, 10%

1 (ACoA aneurysm)
Mesencephalotomy

As per above, plus possible

thalamic impingement

CP due to parietal lesions

(2 pts)

1 (pontine lesion)

11 initial pain disappearances

or abatements, 3 failures,
2 deaths

CPSP (14 pts)

Follow-up: 4 full relapses

(15 mos), 2 partial relapses
(15 mos), 5 long-term good
reliefs
Pain relief (6 mos)

Not available for review

Spinothalamic tract plus reticular formation at midbrain level

CP (2 pts)

Wycis and Spiegel (1962)

Including patients reported in
previous series

Death after 26 h

Torvik (1959)

Open lateral

Thalamic pain (1 pt)

Mesencephalotomies (STT tractotomies and reticulotomies) and other brainstem procedures (coagulations)

Walker (1942)

278

Cause of central pain

Thalamic pain (2 pts)

Laitinen (1988)

CPSP (brainstem) (2 pts)

CPSP (45 pts?)

Brain CP (11 pts)

Thalamic pain (2 pts)

Sampson and Nashold (1992)

Gorecki and Nashold (1995)

Tasker et al. (1991)

Bosch (1991)

Paraplegia pain (1 pt)

(19631985)

CPSP, brain (20 pts)

CPSP, brainstem (7 pts)

Shieff and Nashold (1988)

(includes all patients from
Nasholds previous publications
on this treatment)

Mesencephalotomies and other brainstem procedures (continued)

Author(s)

TABLE 7.1 (continued)

Rostral mesencephalotomy

Mesencephalotomy with/without medial thalamotomy

Caudalis DREZ

STT

2. lesions at inferior colliculus

level

Lesion at:
1. medial lesions at superior
colliculus level

Procedure

Steady pain relieved in 3 (plus

other 3 temporarily) and failed
in 5; intermittent pain relieved
in the only pt who had it; evoked
pain relieved in 3 and unrelieved in 2
0% relief at 1 yr

1 complete relief, 1 partial relief

(448 mos)
50% relief at 3 mos?

14 early pts: 5, 100% reliefs;

6 fair (minimal residual pain,
non-opioids required); 3, 0%
reliefs
12 late pts ( 1 death 1 lost
to follow-up): 7, 100% reliefs;
2 fair, 1 poor (significant residual pain), 2 0% (follow-up:
360 mos)
13 early pts: 5, 100% reliefs,
5 fair; 1 poor; 1, 0%
(1 moribund); 12 late pts:
4, 100%; 3 fair; 2 poor; 3, 0%
?

Efficacy/(follow-up)

Evoked pain more responsive than

steady pain

25% relieved at 3 yrs. Complications

in half, including new dysesthesias
Arm ataxia

Whole neurogenic pain group:

Lesion impinginng on reticular

formation

Gradual disappearance of pain

Unilateral lesions relieved bilateral

pain

4 pts with repeat early surgery,

6 reoperated for late relapse and
1 pain-free after 4 procedures

Notes

279

Bulbar trigeminal stereotactic

nucleotractotomy
Caudalis DREZ

CP, Wallenberg (7 pts)

CP, brainstem

Surgery at T58

1 MS paraplegia pain

1 transverse myelitis with girdle

pain
1 gunshot wound to L1

1 MS

King (1977)

Schvarcz (1978)

Cordotomy
Cordotomy

CP, brain (1 pt)

CP, cord

SCI CP (45 pts)

SCI pain (18 pts)

Turnbull (1939)

Kuhn (1947)

Freeman and Heimburger

(1947)
Davis and Martin (1947)

Cordotomy

C3 anterolateral cordotomy

CPSP (1 pt)

Right, then left cordotomy (C5

and C3)

Stereotactic trigeminal
nucleotomy
Stereotactic extralemniscal
myelotomy

Frazier et al. (1937)

Anterolateral cordotomies (spinothalamic tractotomies)

3 spinal lesions

Surgery at T12S2

1 paraplegia pain at T11

Lippert et al. (1974)

Commissurotomy at T11S1

Surgery at T12S4

2 MS and 1 tabes dorsalis (pain

at L2S2 in both MS and L1S1
in tabes)

Sourek (1969)

Commissurotomy at D1011/
12 in MS and D11L1 in tabes

Midline commissural myelotomy (and stereotactic central C1 myelotomy)

Teixeira et al. (1998, 2003)

Partially effective; 96% relief at

18 mos
1 lasting relief, 2 for 12 wks
with full relapse, all others
failures

Partially effective

Follow-up: 1 yr

Immediate abolition of pain

Relief for 6 mos, then for 2.5

mos until death

0.56 yrs follow-up

Details as of CP not given

Total relief (26 mos, until death)

Much relieved

Relief over 6 mos

All 3 immediately relieved after

surgery. 1 MS case had no
relapse at 1 yr (at 6 mos girdle
pinprick analgesia disappeared). No follow-up in other 2
due to poor general conditions
Over 7 mos relief from 80 to 20%

Orofacial pain <VAS 3 in 85.7%

of pts immediately and at
follow-up (2 yrs)
Failure

(continued)

No relief from previous sympathetic

block

Facial pain treated by gasserian

ganglion alcohol injection

Follow-up: NA

1 pt full relapse in 4 weeks and one

partial relapse in 6 mos. One repeat
procedure

280

Cause of central pain

Anterolateral cordotomy at C2
Higher cordotomy

CP, brain (1 pt)

CP, postcordotomy (1 pt)

White and Sweet (1955)

CP, cord

CP, postcordotomy (1 pt)

SCI CP and root damage (34

pts)

CPSP (1 pt)

Krueger (1960)

Bohm (1960)

Porter et al. (1966)

Waltz and Ehni (1966)

Botterell et al. (1954)

CP after lateral mesencephalotomy (1 pt)

CP, complete SCI (5 pts)

Drake and McKenzie (1953)

Unilateral C2 anterolateral
cordotomy

Bilateral T1T3 cordotomies

Lower bilateral cordotomy

Cordotomy

Spinothalamic tractotomy
at bulbar level
Bilateral high thoracic
cordotomy

Cervical anterolateral cordotomy

CPSP (1 pt)

CP, cord (16 pts)

Pollock and colleagues (1951)

Cervical anterolateral cordotomy

Procedure

Stone (1950)

Anterolateral cordotomies (continued)

Author(s)

TABLE 7.1 (continued)

100% relief; full relapse

at 6 mos

0% for burning pain in the legs

0%

Partially effective

Pain abolition

No pain relief

3 excellent or good pain reliefs

(2, 3, 8 yrs); 1 early failure,
1 late failure (6 yrs)

9 abolition of end-zone pain

(but not diffuse pains)
5 failures
100% relief

2 total pain reliefs

0%

Efficacy/(follow-up)

Previous bilateral thoracic cordotomy. Burning pain

Shooting or electrical-like pain
relieved at least partially in 87%
at 13 mos and 62% at 820 yrs
(cauda pain)
Rhizotomy unsuccessful in cases
subsequently relieved by cordotomy
Burning leg pain not an indication

In each case, pain was sharp,

cramping, stabbing and episodic.
Burning pain in saddle area present
in 2 cases, eliminated in 1
Temporary relief from cortectomy
(see Table)
Electrical and burning sensations,
allodynia
Not available for review

Notes

281

34 (cord CP)

CPSP (1 pt)

Tasker et al. (1992)

Parrent et al. (1992)

SCI pain

CP, cord (1 pt)

CP, cord

Armour (1927)

Davis and Martin (1947)

Freeman and Heimburger

(1947)

Cordectomies

CP, brain (1 pt)

Tabetic crises (3 pts)

Thalamic pain (1 pt)

Removal of a 23 cm cord
segment at T34

Cordectomy at the lower end

of the cord and adjacent cauda
equina (T12L2)
Cordectomy

Cordotomy

Percutaneous cervical
cordotomy

Cervical anterolateral cordotomy

Unsuccessful

0%

Complete pain relief in thighs

and lower abdomen

0%

Steady pain: c. 75% of pts

unrelieved; 20% of pts relieved
2550%; the rest relieved
450%
Intermittent pain: c. half the pts
relieved 450%; c. one third
relieved 2550%; the rest
unrelieved
Evoked pain: half the pts
relieved 450%, one-fourth 25
50% and the rest unrelieved
Results at 1 year

Pain relief

2 reliefs (plus 1 early failure)

1 early failure
Follow-up: up to 12 yrs

5 late reliefs (6 early)

1 early plus 1 late failure

High cervical cordotomy

8 late reliefs (9 early)

Paraplegia pain (2 pts)

High thoracic cordotomy

Paraplegia pain (12 pts)

Partially effective

Tabetic crises (6 pts)

Cordotomy

CP, cord

Sweet (1991)

Joyner et al. (1966) (Freeman

coauthor: see above)
White and Sweet (1969)

Leg pain

War conocaudal injury

(continued)

282

Traumatic T7 total transverse

lesion; pain at T56 (1 pt)

SCI (thoracic gunshot) pain

(1 pt)

SCI pain (4 pts), including 1 pt

with anterior spinal artery
syndrome ASAS

SCI pain (1 pt)

SCI pain (1 pt)

SCI pain (1 pt)

SCI pain (2 pts)

SCI pain (5 pts) with complete

transection

Botterell et al. (1954)

Smolik et al. (1960)

Werner (1961)

Druckman and Lende (1965)

Druckman (1966)

White and Sweet (1969)

Melzack and Loeser (1978)

Cause of central pain

McCarty (1954)

Cordectomies (continued)

Author(s)

TABLE 7.1 (continued)

Pain relief. Follow-up 4 yrs

Cordectomy at various levels

No pain relief

1 partial (1st cordectomy at

T912 abolished part of the

2 unsuccessful (burning pain in

legs, abdomen, buttocks);

Complete pain relief. Follow-up:

18 mos. Persistence of mild
burning in legs
Pain relief. Follow-up: 12 mos

2. Cordectomy up to T11

Unsuccessful in ASAS pt
despite flaccidity
Pain persistence after first
myelectomy. Pain relief after
a 2nd myelectomy
No pain relief

Burning pain in the feet arising

after cordectomy
Pain and spasm relief in 2 pts

Narcotics stopped/follow-up:
6 mos
Annoying girdle pains relieved
6 mos later and occasional
root pain at T5
Girdle pain at lesion level totally
relieved for 8 yrs

Efficacy/(follow-up)

Cordectomy just above trauma

level (T11 vertebra)
Second higher cordectomy 3 cm
above previous one in normal
tissue
Cordectomy above injury
through normal cord
1. Limited cordectomy

Cordectomy

Cord removal from the T10 level

down through conus medullaris
and upper cauda equina

Excision of the damaged cord

up to grossly normal cord
T45 rhizotomy

Removal of the lower 21 cm

of the cord from T5 down to the
conus

Procedure

Sympathetic blocks ineffective

Severe burning pains in legs

End-zone pain. First resection left the

scarred proximal cord stump adhering to the dura
Conocaudal injury. Pain in lower
abdominal and inguinal areas
mild burning in legs girdle pain.
No pain relief from a previous
bilateral T1112 rhizotomy

Notes

283

SCI pain (2 pts)

SCI pain (6 pts), in 5 also

posttraumatic syringomyelia

SCI pain (19 pts), diffuse to

legs in 15

Nashold and Bullitt (1981)

Durward et al. (1982)

Jefferson (1983, 1987)

Cordectomy at T1011 and

T37 ( limited rhizotomy)

1 cm long low thoracic

cordectomy
Cordectomy somewhat above
the area of trauma at T68
(upper level of transection
below the upper level of the
syrinx) in 3 (14 yrs after CP
onset)
Cordectomy at various thoracic
levels (T2 for a C6 lesion, T45
for major injury at T7 plus syrinx,
T1012 for same level injury)
Cordectomy at T11 and/or
below

Cordectomy of tethered cord

Pain relief: 70100% in

14/15 pts (100% in 7/14 pts)
Partial (leg pain abolished,
abdominalgenitalsbuttocks
pain unrelieved) in 1
025% relief

No pain relief of legbuttock

pain in 3

Pain relief of arm pain in 3

Pain relief. Follow-up: 12 yrs

pain for 2 yrs with worsening at

3rd year; 2nd one at T45
ineffective)
2 pain reliefs (paroxysmal
shooting pains in legs abolished
for 11.5 yrs with full relapse;
thoracoabdominal pain abolished by T89 operation with
gradual full relapse by 5 yrs)
Pain abolished

(continued)

In some cases cured of their pains,

there was still severe widespread
cord damage at the upper incision
level

Lesions at/below T11 with episodic,

electric shock/spasm non-burning
pain more likely to respond to
cordectomy immediately, completely
and permanently

T4 fracture; severe pain in legs only

upon head flexion
T12L1 fracture; pain in both legs

284

CP, cord (2 pts)

Pagni and Canavero (1995b)

SCI CP (1 pt)

Richter and Seitz (1984)

Thomas and Jones (1984)

SCI pain (20 pts)

SCI pain (56 pts): end-zone

pain (31 pts); burning
dysesthesic pain (25 pts)

Wiegand and Winkelmueller

(1985)

Friedman and Bullitt (1988)

Tumor CCP (1 pt)

SCI pain (2 pts)

SCI pain (2 pts)

Dieckmann and Veras (1984)

SCI pain (5 pts)

Samii and Moringlane (1984)

DREZ lesions

CP, cord (12 pts)

Cause of central pain

Tasker et al. (1992)

Cordectomies (continued)

Author(s)

TABLE 7.1 (continued)

DREZ lesions (lesions from a

few segments above to a few
segments below)

DREZ lesions

DREZ lesions

DREZ lesions

DREZ lesions

DREZ lesions

Cordomyelotomy (T5SI
myelomeres)

Cordectomy

Procedure

Pain relief end-zone pain:

74% good (100% relief
and/or no analgesics needed
OR residual discomfort not
interfering with daily living

Pain relief (534 mos):

100% in 9, 80% in 1

Good relief

Poor relief

0% benefit

0%

Pain relief: 70100% in 2/5,

5070% in 2/5, <50% in 1/5

Steady pain relief: none in

70% of cases, 2550% in
30% of cases
Intermittent pain relief at 1 yr:
450% in 60% of cases,
2550% in 40% of cases
Evoked pain relief (41 yr):
450% in 80% of cases,
2550% in 20% of cases
Shooting pain/spasms abolished; moderate burning to legs
and perineum lessened

Efficacy/(follow-up)

At follow-up, 10 had maintained

their early postoperative relief and
moved from 80 to 100% relief
Bilateral pain resistant, but 9 of 10
with unilateral pain had good relief

Pain at T23; burning pain in 1,

burning and needles in 1, in
3 unspecified

Long-lasting (10 yrs) pain relief

Notes

285

SCI pain (4 pts)

SCI pain (26 pts)

SCI pain (4 pts)

SCI pain and other pains

(120 pts)

Paraplegia pain (22 pts)

Young (1990)

Tasker et al. (1992)

Edgar et al. (1993)

Rath et al. (1996)

Junctional DREZotomy

Computer-assisted
DREZ lesions
Standard DREZ lesions

DREZ lesions

DREZ lesions (standard

and laser)

DREZ

Extensive DREZ lesions

CP (1 pt), T12 AVM

Kumagai et al. (1990)

DREZ lesions at T35

posterior poliotomy
(LX ablation)

SCI pain (1 pt), trauma

Sweet and Poletti (1989)

DREZ lesions, laser

SCI pain (cauda) (2 pts)

CCP (9 pts)

Powers et al. (1988) (also

includes Powers et al. 1984)

Follow-up: mean 54 mos

End-zone pain relieved in most

who had it

Spinal cord cyst: 5 failures of 7

End-zone pain relieved in 92%

of pts; follow-up: 296 mos
End-zone pain relieved in 58%
of pts
Diffuse burning: 5 failures of 6

2550% relief on evoked pain

present in 2 pts (41 yr)

55% of pts relieved. Follow-up:

up to 5 yrs
83% of pts with cauda equina
lesions relieved
No effect on steady pain

(continued)

93% had diffuse pains and/or

sacral pain

Midline pain, especially in midlumbar area or genitalia, unrelieved;

end-zone pain benefited

Not available for review

Midline (perineal, scrotum) pain:

relief in 0/3

Complete relief of thoracic

end-zone pain and
coccygeal/foot pain for 3 mos.
At 13 mos, 450% relief
0% of diffuse bilateral pain from
lower abdomen downwards
50% relief at 1130 mos

Below-level pain: relief in 2/8

Follow-up: 463 mos

End-zone pain in 4: all relieved

0%

activities), 6% fair (still requiring

some analgesics), 20% no
result
Pain relief diffuse dysesthetic
pains: 20% good, 12% fair,
68% no results
5 successes, 4 failures

286

Pontine CPSP (1 pt)

Sampson and Nashold (1992)

Falci et al. (2002)

SCI CP (1 pt)

Prestor (2001)

SCI pain (41 pts), generally at

T10L1, but 6 cases at T49

Syrinx CP (6 pts)

SCI (44 pts)

Sindou et al. (2001) (includes

all previous papers of Sindou
on this procedure)

CP, mesencephalic AVM (1 pt)

Conocaudal pain (39 pts)

Cause of central pain

Nashold and Pearlstein (1996)

(includes all previous papers of
the Dukes group on this
procedure)

DREZ lesions (continued)

Author(s)

TABLE 7.1 (continued)

DREZ lesions guided by multiple

electrophysiological techniques

Junctional DREZ

Radicellotomy

Caudalis DREZ

DREZ lesions

Procedure

Good (16.7%)
at 648 mos
Group A (9 pts): 100% relief
in 56% of pts (50100% relief
in 78%); follow-up: 67 yrs
Group B (32 pts): 100% relief in
84% (50100% relief in 88%);
follow-up: 16 yrs
End-zone pain (present in 6 of
32): 100% relief in all. Belowlevel pain (present in 26 of

Excellent relief (83.3%)

0%

Long-term good results in

68% of pts

450% pain relief in 14/16 pts

(6 mos7 yrs)

100% relief 2 days later

over 4 yrs
50% relief 8 days later (death
4 mos later during surgery)

Pain relief at a mean of 3 yrs:

good (no analgesics required) in
54% of pts; fair (nonnarcotics
still necessary, but pain not
interfering) in 20% of pts
100% relief in 35% of pts
at 10 years

Efficacy/(follow-up)

Evaluation: telephone interview and/

or outpatient evaluation (VAS/verbal
scales)

15% of repeat surgeries

4.7% of pts developed a new
permanent pain of low intensity
(VAS 13)

Below lesion pain not favorably

influenced, particularly
perineosacral; radiculometameric
pain responsive

Narcotics down from 90% of pts

to 12%. Conocaudal pain relieved
in 60%
Best results in electric shock pain
and end-zone pain
Facial pains abolished in
40% at 10 yrs

Notes

287

SCI pts (complete/incomplete)

(11 pts)

Rogano et al. (2003)

DREZ lesions

DREZ lesions

CP (including syringobulbia)

CPSP (1 pt)

SCI pain, conocaudal, burning

BCP (1 pt)

CPSP (1 pt)

SCI pain

CP (face only) (1 pt)

CPSP (1 pt)

Garcin (1937)

Frazier et al. (1937)

Slaughter (1938)

Turnbull (1939)

Hecaen et al. (1949)

Pollock et al. (1951b)

Spiegel et al. (1952)

Bonica (1953)

Good results for 2 mos, relapse

and again benefit from a new
block

Benefit

Stereotactic mesencephalotomy
Repeated paravertebral
pantocaine blocks and later
subarachnoid alcohol block

Failure

Failure

Failure

Failure

Trigeminal rhizotomy

Sympathectomies

Stellectomy.

Sympathetic block

Sympathetic block

Relief

Good results

Goodtotal control of pain

Retrogasserian neurotomy
Bilateral cordotomy, alcohol
injection of the gasserian ganglion and radicotomy of C23
Sympathectomy

Failure in 2 cases

Alcohol injection of the gasserian ganglion

Spinal rhizotomies, peripheral blocks, sympathectomies and sympathetic blocks

SCI pain (T9L4) (26) pts

Spaic et al. (2002)

32 pts): 100% relief in 81% of

pts (50100% relief in 85%)
Thermal pain (burning and
similar), steady pain and diffuse
infralesional pains: 0% longterm relief.
Shooting, cutting, stabbing,
sharp, cramping, constriction,
throbbing end-zone pains:
100% relief in 70% of pts
and 450% relief in 20%
at 1350 mos
VAS from 9.7 to 1.9: end-zone
pain only

(continued)

Peripheral blocks eliminate normal

afferent stimuli

Subsequent unsuccessful cortectomy and lobotomy

Review of cases reported by Foix,

Foerster, Schaefer, Ravina and
Haguenauer, Parker

288

Cause of central pain

Procedure

Sympathectomy

Alcoholic injection of the

thoracic sympathetic chain

CPSP (6 pts)

SCI pain (5 pts) with complete

transection
1. CP, brainstem (no confirmatory CT scan)

Hannington-Kiff (1974)

Melzack and Loeser (1978)

2. CPSP
3. CPSP

CPSP, brainstem (1 pt)

2. 50% 1 h relief
3a. Pain and allodynia much

1. Complete permanent relief

1. Guanethidine block (leg)

2. Sympathetic chain block

3a. Guanethidine block 1

Failure

6/6 pain relief

0% relief

Failure

Failures

Failures

Failures

Failure

Failure

Sympathetic blocks

Peripheral blocks (local anesthetic, alcohol) in trigeminal

branches
Guanethidine sympathetic block

Sympathectomies

Nashold and Wilson (1970)

Loh et al. (1981) (likely

includes almost all pts
previously published in Loh et
al. (1980) and Loh and Nathan
(1978), where apparently a

Sympathetic lidocaine block

Alcohol injection of III branch of
trigeminal n. and later gasserian
ganglion
Retrogasserian rhizotomy

Rhizotomy

White and Sweet (1969)

CCP, postcordotomy (1 pt)

CPSP (1 pt)

CPSP, Wallenberg (1 pt)

Waltz and Ehni (1966)

Sympathectomy

CP (several pts)

SCI pain

Porter et al. (1966)

Failure
Good pain relief

Stellectomy
Stellectomy

Trigeminal rhizotomy

CP, brain (1 pt)

Campanini and De Risio (1962)

Failure

Efficacy/(follow-up)

Retrogassearian neurotomy

CP, syringobulbia

CP, brain (1 pt)

Rowbotham (1961)

Spinal rhizotomies, peripheral blocks, sympathectomies and sympathetic blocks (continued)

Author(s)

TABLE 7.1 (continued)

All pts had burning pain plus

several kinds of allodynia/
hyperpathia
No placebo injections!
Only pts with hyperpathia-allodynia

Placebo (saline) or bupivacaine

ineffective

Notes

289

further tumor CCP case and a

postcordotomy CP case were
also included)

7. CP, multiple sclerosis

6. CP, multiple sclerosis

5. CCP (Schneiders syndrome)

4. CPSP (negative CT scan)

7a. Guanethidine block 1

6f. Iontophoretic guanethidine 1

and 2

6e. Guanethidine block 5

6d. Guanethidine block 4

6c. Guanethidine block 3

7a. Slight effect on burning and

allodynia for 2 h

6b. Burning abolished,

allodynias abolished or much
improved for 60 h
6c. Burning 75% reduced,
allodynias abolished or much
improved for 20 h
6d. Burning 50% reduced,
allodynias improved or much
improved for 2024 h
6e. Burning abolished,
allodynias abolished or
improved for 46 h
6f. Pain reduced in fingers

6b. Guanethidine block 2

6a. Guanethidine block 1

5b. Guanethidine block

5a. 50% relief of burning and

allodynia improved for 20 h
5b. Slight relief of pain, allodynia
slight or moderate relief for 24 h
6a. Burning improved (?),
allodynias removed or much
improved for 46 h

4b. Complete relief for 12 h

improved for 5 days

3b. No effects
3c. Pain almost gone, allodynia
much improved for 40 h
4a. Complete relief for
2.548 h

5a. Sympathetic chain block

4b. Guanethidine block

4a. Sympathetic chain block

3b. Guanethidine block 2

3c. Guanethidine block 3

(continued)

Stellate block: local anesthetic; IV

guanet: 15 mg in 30 ml saline

Pain relief beyond area of block in

2 cases. Case 4: stellate block also
blocked leg pain; case 3: neckear
pain relieved, plus shoulder pain

No effects on sensibility of normally

innervated regions, but pts may
notice an area of diminished sensibility and numbness restricted to the
territory of the damaged nerve

Also, guanethidine infusion did not

completely block sympathetic
control of digital blood vessels.

290

Cause of central pain

Procedure

Yamamoto et al. (1997)

Syringomyelia CP

D. Long (comment to Milhorat

et al. 1996)
Milhorat et al. (1996, 1997)

CPSP, thalamic and suprathalamic (39 pts)

Syringomyelia CP (1 pt)

Syringomyelia SCI CP (2 out

of 15 pts)

CCP (5 pts)

8. CP, cervical astrocytoma

Tasker et al. (1992)

Loh et al. (continued)

1. Sympathetic block with

10 ml of 0.25% bupivacaine,
then stellate ganglionectomy
2. Similar blocks
sympathectomy
Stellate ganglion block

8d. Iontophoretic guanethidine

1 and 2 in phalanges
Rhizotomy (L4, T12L1, L12
bilaterally, intercostal nerves)
Sympathectomy

8c. Guanethidine block 2

8a. Left sympathetic chain

block
8b. Right guanethidine block 1

7c. Guanethidine block 3

7b. Guanethidine block 2

Spinal rhizotomies, peripheral blocks, sympathectomies and sympathetic blocks (continued)

Author(s)

TABLE 7.1 (continued)

2. Relief
100% relief 22 mos later
Failure

1. Prolonged relief, then 100%

relief at 5 mos

7b. Burning halved, allodynia

much improved or abolished for
60 h
7c. Burning halved, allodynia
modestly improved for 3 h
8a. Burning almost abolished
and allodynia improved for 1 h
8b. Burning halved, allodynia
improved for 1 h
8c. Burning halved, allodynia
improved for 1 h
8d. Tenderness improved,
allodynia gone
Pain relief in 2/5 (C4, T512)
(hyperpathia only)
Most failures

Efficacy/(follow-up)

Stellate ganglion block with 10 ml of

0.5% mepivacaine and to cervical or
lumbar epidural block with 5 ml of
0.5% mepivacaine

Rhizotomy only transiently effective

for steady pain and then worsening

Notes

Pathophysiology: Human Data

between individuals, but also between hemispheres in an individual (Penfield above).

Anyway, SI cortectomies have a better track record than, for instance, frontal
operations, including cingulectomy/cingulotomy and focal lesions of SI can indeed
abolish CP in a somatotopographical fashion (Canavero et al. 2001). As suggested
by electrophysiologic data (section on neurophysiology above), cortical stimulation
studies (Chapter 6) and sudden disappearances of CP following lesions ipsilateral
to pain (section on reports of sudden disappearance of CP below), some failures
of cortectomies and thalamotomies to relieve CP but also cases of CP with apparent
total destruction of SI can simply be chalked up to lesioning the wrong side, as
the corticothalamic loop we posit at the basis of CP has shifted ipsilaterally to pain
(Chapter 8). After SI damage, input may also be rechanneled to surviving areas of
SI or other sensory zones (e.g., SII) (see Bittar et al. 2000). In sum, SI is involved
in the mechanism of CP.
A considerable amount of evidence suggests that SI has a pivotal role in sensory
discrimination/localization of pain (reviewed in Willis and Westlund 2004).
The inconsistency of results of early lesion and functional imaging studies (only
half reported SI activation) on the role of SI in pain processing has been explained:
the probability of obtaining SI activation appears related to the total amount of body
surface stimulated (spatial summation) and probably also by temporal summation
and attention to the stimulus (see Schnitzler and Ploner 2000). Anatomically,
SI consists of four cytoarchitectonically defined areas, each with a representation
of the body surface; unlike the tactile modality (reviewed in Iwamura 1998), pain
processing appears to be less hierarchically organized, with BA1 as main focus of
nociceptive processing with nociceptive neurons clusters in layer IIIIV (Schnitzler
and Ploner 2000; Willis and Westlund 2004).
Unlike all other cortical areas (including SII and ACC), SI is the only one with
a clear somatotopic organization on neuroimaging studies (Coghill et al. 1999),
an essential pathophysiologic consideration. Actually, SI does not truthfully map the
body surface (somatotopic homunculus) on all occasions but, depending on the
stimulus, may represent an internal brain image that is linked to subjective
perception, rather than to objective sensory input, being activated in a manner that
corresponds to the perceived stimulus. Thus, representations on SI may both reflect
integrated higher brain functions and simple topographic representations of physical
stimuli detected by the periphery. The degree of SI activation enabling emergence of
a perceived image is related to the type of information that generates the illusion.
In many cases, the image of the world within the brain is congruent with neither the
real nor the perceived world (Eysel 2003).
SI may be directly involved in elemental awareness with a role of 40 Hz coherence
in conscious perception at 150300 ms (Meador et al. 2002) (but not at 40 ms; Preissl
et al. 2001). However, conscious awareness of a stimulus location on the body likely
involves the interaction of other brain regions along with SI, including BA40 (inferior
parietal lobule) and portions of the dorsolateral prefrontal cortex (DLPFC).
Moreover, pain is highly intrusive, attention-grabbing and is widely distributed
(SI, PCC, DLPFC, ACC).
Human evidence indicates that SII is involved in recognition of the painful
nature of the stimulus (particularly if moving) and may play an attentional role

291

292

Central Pain Syndrome

(hence, its bilaterality of activation), but is clearly not essential for stimulus
localization/discrimination (see Schnitzler and Ploner 2000; Fujiwara et al. 2002);
in humans, it receives few fibers from Vc (Kaas 2004). Unlike tactile input,
noxious input simultaneously activates SI and SII (see Schnitzler and Ploner 2000;
see also Hobson et al. 2002). First pain is particularly related to SI activation, second
pain to ACC activation; both are associated with SII activation (Ploner et al.
2002). Actually, SII and the (right) posterior insula may be considered as a unique
structure (and cannot be separately resolved by present day PET) (see also Frot and
Mauguiere 2003). The insula may integrate pain-related input from SII and the
thalamus with contextual information from other modalities before relaying
this information to the temporal lobe limbic structures (pain-related avoidance
memory/learning) and to autonomic stations (amygdala, brainstem, etc.). Patients
with insular lesions recognize a stimulus as painful, but exhibit absent or inappropriate affective responses (Berthier et al. 1988), as early stages of affect are mediated
in the insula.
The posterior parietal cortex (PPC, BA5-7) may play a role in conscious pain
perception and body awareness (Witting et al. 2001) and in the initial stages
of cortical motor planning (Driver and Mattingley 1998). PPC receives input from
SI and SII, while the DLPFC and the PPC are the most densely connected areas
of the association cortex (and may actually process attentional-orientation toward
incoming sensory input). Lesions to PPC produce multisensory (body schema)
neglect syndromes. Nonetheless, a role in CP is questioned by Hoogenraad and
colleagues (1994), who described a 46-year-old man with ischemic infarction of the
right parietal cortex following carotid dissection and, among others, left hemianesthesia with almost complete loss of all sensory modalities. MRI disclosed an infarction involving the posterior part of the postcentral (SI), supramarginal and angular gyri
plus inferior and superior parietal lobe. Over the next month the patient was unaware
of his left arm, had no feeling in the arm, could not use it, but when he saw the arm
being approached by someone it would suddenly move sideways as if it had been
stung; simultaneously, he experienced a burning pain. The involuntary withdrawal
movements of his left arm were so embarrassing that he tied it to his belt. Eight
months later, with eyes closed, he showed loss of superficial sensation (pain and
touch) in the left side of his body, more severely in the arm than in the leg, trunk and
face, the distal parts of the extremities being affected most. No delayed pain reaction
occurred. There was also complete loss of postural sense, which resulted in sensory
ataxia and pseudo-athetoid movements. Vibration was not perceived. There was lack
of awareness of the left half of his body and inability to move his left hand and fingers
without visual control. With his eyes open and his gaze directed at his left hand, the
patient was able to open and close the hand very slowly. There were no sensory
abnormalities on the right side of his body. On seeing that the left part of his body
was approached for sensory testing, the patient invariably made a brisk withdrawal
movement; at the same time he felt a burning pain that was accompanied by
grimacing. On moving about, an incidental contact that was not anticipated did not
result in pain and withdrawal. When the patient himself approached his left arm
with his right hand there was neither pain nor withdrawal (suggesting that attention
activates CP).

Pathophysiology: Human Data

b.

( L O B O T O M Y , T O P E C T O M Y , C I N G U L EC T O M Y /
Unlike other chronic pains (Bouckoms 1989), results
are generally disappointing for CP. In rare cases in which it was deemed effective,
the pain was simply less distressing and bothersome (pain indifference), the patient
less anxious or depressed by pain; spontaneous complaints about pain are diminished
and a patients ability to appreciate the meaning of the pain may be disrupted.
According to Turnbull (1972), bilateral cingulotomy alone is ineffective when pain
is caused by a major organic disease (p. 962), including CP.
Bilateral cingulotomy/capsulotomy (but also some psychiatric conditions) result
in decreased pain tolerance and hyperphatic-type responses to acute painful stimuli
following frontal surgery (e.g., Davis et al. 1994; Talbot et al. 1995). This is the reverse
situation expected from some theories (Craig 1998), in which interruption of the
thalamocingular path or destruction of cingular areas may actually relieve CP.
Contrary to some speculations (Pattany et al. 2002), frontal lobes are not essential
to CP generation. However, prefrontal activity may lead to an increased salience
of pain at the cost of other cognitive and emotional behavioral abilities, with pain
constantly interfering with attention to other tasks.
In humans, the anterior cingulate cortex (ACC) may be divided into a caudal
region, showing increased activity during pain per se (from STT input), an
adjacent part preferentially involved in general attention (alerting/orienting attention
[at 125 ms] and [escape] response competition monitoring [at 200 ms] to pain;
Dowman 2002) and a rostral region involved in pain affect (i.e., unpleasantness
of pain). Rostral ACC (and/or underlying cingulum) tonically suppresses pain,
with opposite effects on pACC and insula (reviewed in Petrovic and Ingvar 2002),
but chronic pain engages both ACC and mid-cingulate cortex (MCC) (which also
includes cingulate motor areas) (Vogt et al. 2003); differential involvement of MCC
in pain may result in different outcomes in cingulotomy analgesia. Interestingly,
differences between physically and psychologically induced pain may be quantitative
rather than qualitative, with a role of rostral/perigenual ACC and pericingulate
areas in source monitoring (Raij et al. 2005).
The significant involvement of CC in pain processing may be an evolutionary relic
from a distant past when the prefrontal neocortex had not yet evolved and hippocampus, cingulate cortex, cingulate and brainstem motor areas/nuclei and amygdala
represented the highest order cognitive, afferent and efferent levels (McCrone 1999).
In humans, the assembly of information and motor-autonomic response to a painful
experience may depend largely upon the evolutionarily late PFC and its extensive
output to multiple brain sites, with a particularly important role for late cognitively
driven stages of pain affect and for the sharp consciousness of a mental event
(McCrone 1999). According to Freeman and Watts (1950), the frontal lobes are
important structures, not so much for the experiencing of pain as for the evaluating of the sensation, the estimation of its significance in terms of the self and
of the future. However, this network plays clearly no primary sustaining role in
chronic CP.
Finally, Guiots group is said to have temporarily relieved CP by bilateral ablation
of BA6 (Garcin 1968), but stimulation in these areas never provided a benefit (one
personal case plus others from a Japanese group; see Chapter 6).
FRONTAL

(PSYCHIATRIC)

293

C I N G U L O T O M Y , L E U K O T O M Y ):

S UR G E R Y

294

Central Pain Syndrome

c. H Y P O T H A L A M O T O M Y A N D H Y P O P H Y S E C T O M Y : Amano (1998) concluded that,

unlike cancer pain, posteromedial hypothalamotomy is not effective at all for
neurogenic pain, including CP, thus disproving the theory of Spiegel and colleagues
(1954) and Spiegel and Wycis (1962) of CP arising from diversionary impulses on the
hypothalamus. Interestingly, no postoperative sensory deficit is apparent; chronic
cancer pain disappears, but pain can still be induced by pinprick. This dissociation after
posteromedial hypothalamotomy is similar to that seen after medial thalamic lesions.
Why pituitary lesions can temporarily allay some CP patients is a matter of
speculation. While a placebo effect cannot be excluded in reported studies, according
to Levin (1988)
Pain relief may result from excitation of central pain-suppressor mechanism by means of
either a humoral agent distributed by the CSF . . . or by a direct neural stimulus.
Hypophysectomy . . . either eliminates a hormone responsible for pain augmentation
produced by the pituitary or induces (possibly by elimination of feedback suppression)
a neural or humoral response, originating from the hypothalamus, which is responsible
for pain suppression.

This humoral factor could be arginine-vasopressin, with involvement of a

hypothalamo-thalamic antinociceptive pathway (Fujita and Kitani 1992) or corticotropin releasing factor (CRF), a peptide secreted from the hypothalamus
throughout the brain with significant analgesic effects by the IT route.
d. T H A L AM O T O M I E S : The literature is for the most part too old to be significant and
many series of thalamotomies did not differentiate results according to pain category
and are not available for discussion; most are pre-CT and MRI. Importantly, there
is poor agreement on thalamic nomenclature among series. With older technology,
it is difficult that lesions may have been limited to Vcpc and also other nuclei
are difficult to evaluate.
Thalamotomies for CP aimed at lesioning the entrance point into the thalamus of
quinto and spinothalamic pain fibers, limitans nucleus, Vc or nonspecific nuclei
(CM-Pf, CL, DM, pulvinar and anterior nuclei) were believed to involve the spinoreticulothalamic (polysynaptic) pain pathways or thought to modify the emotional
response to pain. Paradoxically, therapeutic lesions in Vc resulted in CP (White and
Sweet 1969; Siegfried and Krayenbuhel 1972). Cassinari and Pagni (1969) concluded
that only large thalamic lesions centered on CM-limitans-CL nuclei would completely interrupt spinoreticular pathways (partial lesions would be only temporarily
effective by a temporary suppression of hyperactivity of thalamic or cortical neurons,
for lack of facilitation). Lesions centered on Vc always encroached on the nuclei
of the diffuse projection system of the thalamus immediately close by, and this might
have either promoted or limited CP onset. Mazars (1976, p. 141) stated that all
posterior thalamotomies are followed, after a more or less long time, by CP. Basal
thalamotomies, placed above the midbrain at the base of the medial thalamus,
extended laterally to interrupt both specific and nonspecific pain afferents, and
exactly enclosed Vcpc: results have been similar to other sites. Independently of
the targeted nuclei, initial results of thalamotomies are positive in most cases,
with immediate relief of CP after Vc, CM and pulvinar lesions in some patients

Pathophysiology: Human Data

(see Table 7.1). Results appear to be modestly better (and complications lower, with
no or little sensory loss) with medial (particularly bilateral) than with Vc
thalamotomies (see also Tasker 1990). Bilateral medial lesions, though, increased
the risk of cognitive impairment, by interfering with attentional processes. Few CP
patients appear to have benefited in the long term. The great variability of response,
relapse rate of pain (up to 50%), non-negligible operative mortality, dysphasia and
severe dysesthesias make stereotactic thalamotomy a poor option for CP. Bilateral
lesions produced many more complications and deaths and bilateral extensive
destruction of thalamus is incompatible with life; severe, permanent complications
and deaths have been reported with all thalamotomies. Interestingly, some unilateral
lesions relieved bilateral pain.
Recent image-guided series provide some additional data. Jeanmonod and
colleagues (1996, 2001) found 50100% improvement in 40% of CP much less
than for PNP at 2 years, in line with the experience of Tasker (1990) and Young and
colleagues (1995), after medial thalamotomies (see also Ohye 1998). The lesions
centered in CL, where most bursting units were found, revealed themselves to be
the most efficient. Next, in descending order of efficiency, came Pf, PO, PuO and
PuM nuclei. Results after lesions in CM and midline nuclei were the least efficient.
However, steady pain with thermal qualities proved the most resistant pain profile
than intermittent pain and allodynia, deep (proprioceptive) pain more resistant than
superficial pain. Magnin and colleagues (2001) observed that in neurogenic pain
(including CP) CL stimulation leads to paresthesia, in motor disorders to motor
reactions and in psychiatric disorders to emotional feelings, i.e., CL is a supporting
nucleus, not specific to CP. It should be noted that pulvinotomy, like medial
thalamotomies, can reduce chronic, but not acute, pain (Richardson 1974).
Tasker (2001a) concluded that there may be a place for medial thalamotomy for
evoked-intermittent pains. On the other hand, Ohye (1998) found Vim thalamotomies effective for deep pain only in about 40 CP cases. He also concluded that
CM-Pf used as a target in the past may have been the wrong target (Ohye 1990; but
see Weigel and Krauss 2004). This is interesting, as old series did not distinguish
the various components of CP sufficiently. Excellent results for CP have been
reported after pulvinotomy by some (Yoshii et al. 1980; Laitinen 1988), but these are
difficult to analyze (Tasker 1990).
Taken together, available data suggest involvement of several thalamic nuclei in
the genesis of CP. Certainly, unlike medial lesions, Vc lesions add to denervation,
perhaps resulting in less long-term relief due to shift of the CP generator contralaterally (see Chapter 8). VMpo plays no role in the genesis of CP (Montes et al.
2005).
The puzzling efficacy, at least in the short term, of lesions of different nuclei may
be explained by invoking current anatomical concepts. Cortical areas can speak
to each other through higher order thalamic nuclei, with one thalamocortical (TC)
pathway reporting to its own cortical area the major (layer 5) output of another
cortical area (i.e., higher order TC cells have a role in corticocortical communication)
in tonic mode. Through their layer 6 corticothalamic (CT) connections, they can in
turn modify the report of a cortical output, as this is passed through the thalamus,
by promoting burst or tonic mode. Higher order nuclei receive from layer

295

296

Central Pain Syndrome

5 pyramidal cells about the cortical output (versus first order nuclei). Within each
sector of the reticular nucleus (TRN), cortical areas with the same spectrum of
function (e.g., pain) may influence each other through the action of TRN on sensory
or associative thalamic nuclei. Cortical areas receiving thalamic afferents from higher
order relay nuclei may well be dominated by that input, rather than by other direct
cortical connections (explaining, for example, pulvinotomy effects on CP: pulvinar
is a higher order nucleus projecting to SI, but without STT input). Neurons in
separate somatosensory nuclei of the dorsal thalamus influence (excite or inhibit) one
anothers activity through the TRN (Crabtree et al. 1998), further contributing to
efficacy of different thalamotomies.
Interesting cytoarchitectonic data strengthen the concept. The spread of coherent
activity across ensembles of cortical neurons has traditionally been ascribed to
intralaminar nuclei (Castro-Alamancos and Connors 1997), but in fact this 40 Hz
synchronization can be the sole result of a matrix of calbindin-immunoreactive
(CAL) neurons present in all thalamic nuclei and projecting diffusely to superficial
layers of several adjacent cortical areas (Jones 2001). In some nuclei, a core of
parvalbumin-immunoreactive (PA) neurons is superimposed upon the matrix.
Core neurons project in a topographically ordered fashion to middle layers of the
cortex in an area-specific manner. Matrix neurons, recruited by corticothalamic
connections, can disperse activity across cortical areas and thalamic nuclei. Their
superficial terminations can synchronize specific and nonspecific elements of the
thalamocortical network in coherent activity (perhaps also explaining bilateral
recruitment during allodynia). Subcortical inputs too (e.g., the STT) adhere to this
scheme, being less precise to matrix and more focused on the core. Thus, after
a population of cortical cells is activated by whatever stimulus, it feeds back onto
the matrix cells of its thalamic relay neurons, engaging, via the diffuse projections
of the matrix cells, other adjacent populations of cortical cells. These cells (layer VI)
in turn would feed back to the matrix cells of their thalamic relay nuclei, and so on,
forming links between distant neurons. Multiple thalamic nuclei could be recruited by
corticothalamic fibers returning from the first area to nuclei other than that from which
that area receives its principal thalamic input and might be a key element in binding
together the activities of multiple cortical columns in the generation of a sensory percept.
If sufficiently widespread, it could provide a basis for interactions between distant
cortical areas in uniting perception with planning strategies for action, but also
explain effects of thalamotomies.
The terminations in superficial layers of matrix neurons, together with those of
PA cells in middle layers, form a coincidence detection circuit. The vertical integration of coincident matrix inputs to apical dendritic branches of cortical pyramidal
cells in upper layers and of PA core inputs to their dendrites in middle layers
should promote oscillatory activity in these cells. This would be reinforced by the
projections of cells in layers VVI back to the thalamus, first engaging core and
matrix cells that are topographically related to an activated set of cortical columns,
but soon spreading across more widespread areas of thalamus and cortex. In this way,
transient links would be formed between discrete populations of cortical and thalamic
cells with different relationships to a cognitive event. Attention would modulate this
synchronized neuronal activity and affect intensity of CP.

Pathophysiology: Human Data

e. M E S E N C E P H A L O T O M I E S : These have been performed both to interrupt the STT

or the reticular formation. At rostral mesencephalic level, the medial lemniscus,
neospinothalamic tract, reticulothalamic tract and PAG lie contiguously adjacent
to one another (from lateral to medial, respectively). Since STT lesions but not
coagulation of the termination site of the paleospinothalamic path triggered
new CP (Cassinari and Pagni 1969), most surgeons treating CP attempted larger
medial lesions impinging on the reticular formation, thus including the paleospinoreticulo-thalamic pathways (often combined with medial thalamotomy). The larger
lesions appeared more effective for relief of central dysesthesia (Nashold et al. 1969).
However, Tasker (1989), reviewing 92 published protocols of patients with CP/PNP,
showed that only 27% gained satisfactory long-term relief, from mesencephalotomy,
with several complications and operative deaths. Laitinen (1988) concluded that
mesencephalotomy has no place in the treatment of chronic pain. The efficacy
of this approach is no better than that of nonspecific thalamotomies, but side effects
are more frequent and more serious and Bosch (1991) also concluded against the
use of mesencephalotomy in CP. There are more than 70% postoperative dysesthesias
after open and 1520% after stereotactic mesencephalotomies, with 510% mortality
in stereotactic series (Tasker 1989). However, Amano and colleagues (see Table 7.1)
achieved complete or near complete long-term relief in almost two thirds of their CP
patients, with no postoperative dysesthesias or deaths, by aiming only at the reticular
formation (pure rostral medial reticulotomy). Their target was located at the
border between the PAG and the medial end of the mesencephalic reticular
formation (RMR) at the level between the superior colliculus and the posterior
commissure (Amano et al. 1980). The pretectal area was avoided by burring at
30% of glabella-inion distance. Microrecording showed nociceptive neurons in the
RMR, characterized by large RFs and delayed firing in response to pinprick
stimulation. High-frequency stimulation produced severe pain mostly contralateral to
the side of stimulation in a very restricted area. Similar results were reported by Shieff
and Nashold (1988; see Table 7.1). These latter authors observed how CP resolved
gradually, never suddenly (unlike subparietal lesions), after mesencephalotomy
(Amano et al. did not discuss this point); also, unilateral lesions relieved bilateral
pain. In any case, somatotopographical constraints exclude a primary role of the
reticular substance in the genesis of CP, since the spinoreticulothalamic system has
very large and/or bilateral RFs, while CP is generally unilateral (references in Willis
and Westlund 2004). Thus, the reticular formation may be involved in modulating
a rostral generator and/or conscious experience of CP (Chapter 8).
f. O T H E R B R A I N S T EM P R O C ED U R E S : Trigeminal nucleotomy and DREZ exert their
effect by interruption of the intranuclear polysynaptic trigeminal pathway.
Midline myelotomy may act by interrupting multisynaptic midline pathways and
pontine lesions spinoreticular pathways.
The open caudalis DREZ operation has been successful in relieving the facial pain
of pain resulting from damage to the trigeminal pathways in the brainstem.
Pain due to brainstem involvement was reduced in 67% of cases (Nashold and
Pearlstein 1996). However, the number of patients receiving this and similar
interventions is too small to afford conclusions.

297

298

Central Pain Syndrome

g. A NT E R O L A T E R A L C O R D O T O MI E S ( S P I N O T H AL A M I C T R A C T O T O M I ES ): According to
Joyner and colleagues (1966), 103 reported cordotomies have successfully relieved
paraplegia pain (CP not broken down), and only 27 of the 154-strong group were
unrelieved. On the other hand, White and Sweet (1969) reported that, despite
an initial 56% incidence of pain relief in paraplegics, only 33% have remained
pain free in the long term. Low cordotomies have been much less successful than
higher ones, all at the expense of significant sensory loss. Davis and Martin (1947)
found cordotomies ineffective in several cases of CCP.
The pain that responds to cordotomy is not the steady pain. White (1963) stated
that when the spinal cord is involved rather than its sensory roots, spinothalamic
tractotomy, or even a complete myelotomy, is not likely to eliminate pain in the back
and legs. Botterell and colleagues (1954) stated: in complete lesions . . . burning
pain has proved a problem difficult of solution in cases of injury to the . . . spinal
cord, but by contrast, jabbing, shooting, crampy, gripping, colicky and vice-like
pains, have been regularly relieved by satisfactory bilateral tractotomy (i.e., open
cordotomy). Porter and colleagues (1966) stated that cordotomy has . . . no
effect . . . on the frequently encountered burning pain in the lower extremities
(in traumatic paraplegia). According to Lipton (1989), cordotomies should not be
used (for denervation pains) because when pain returns it may have dysesthetic
qualities and the patient is worse off than previously. Rosomoff (1969) considered
cordotomies futile for CCP and found a high incidence of associated dysesthesias
in this group. Tasker (1990) too stated that long-surviving cord CP patients often
relapse, or new pains emerge and/or the analgesic levels achieved by cordotomy fade
with time. By interrupting the spinothalamic fibers, this obviously sets the stage for
further later different pains (although it was suggested that bilateral cordotomies may
lessen this risk). According to Tasker and North (1997), postcordotomy dysesthesias
typically take time to develop and occur in 11.5% of cordotomized patients, in
about 4% being long lasting and severe. Other series provide higher figures (620%),
but data in many series are difficult to interpret. Tasker (1997) operated on 23 CCP
patients with percutaneous, plus 8 with open cordotomy. Pain recurred in 8 after 1
21 years with gradual fading of analgesia. Repetition of cordotomy in 6 restored the
level of analgesia in all, but pain relief was recaptured in only 3. He (Tasker et al.
1992) relieved spontaneous pain in 27%, intermittent spontaneous pain in 86%
and evoked pain in 75% of his SCI CP cases, showing how intermittent/evoked pains
were dependent upon transmission in STT paths. Thus, STT-tomies may relieve
some cases of CCP (and exceptionally BCP); in the majority, CP relapsed shortly
contralaterally (White and Sweet 1969), ipsilaterally (Bowsher 1988) or bilaterally
(Graf 1960).
Thus, MS CP, CPSP, postcordotomy burning pain and pain due to scarring
of the upper thoracic spinal cord are poorly responsive to anterolateral cordotomies
(but also cordectomies and traditional DREZ lesions), with some exceptions
(single cases of, e.g., Botterel et al. 1954; Davis and Martin 1947; Pollock et al.
1951a). Hyperactive spinal cells may feed the brain generator; reduction of this
bottom-up barrage in some patients (obtained by cordotomy, cordectomy and
DREZ lesions) may at least transitorily interfere with supraspinal mechanisms,
as discussed above.

Pathophysiology: Human Data

h. P O S T E R I O R C O R D O T O M I E S A N D C O M M I S S U R O T O M I E S : Patients with clear-cut CP

are not on record or are just a handful.
i. C O R D E C T O M I E S : Cordectomies relieved the same types of pain that respond to
cordotomy and DREZ surgery.
j. D O R S AL R O O T E N T R Y Z O N E ( D R E Z ) L E S I O N S : The first drezotomy
(Radicellotomie) for paraplegia pain was done in December 1972 by Sindou,
consisting in incomplete section of the dorsal root plus a small 1.52 mm section cut
into the lateral portion of the dorsal horn. Nashold did his first DREZ lesion in
September 1974, with the aim to destroy the damaged dorsal horns, where hyperactive secondary nociceptive neurons were thought to generate the pain state.
The DREZ operation in the paraplegic is generally done bilaterally (unilaterally in
case of one-sided pains), beginning at the level of the traumatic transection of
the spinal cord and extending rostrally over the next three dorsal roots and caudad
over two levels; laminae I through V are ablated. Complications are common and
include a rise in sensory level in all, partial or complete loss of pinprick and
light touch sensation in 7080% of patients, motor deficits (up to 14%), CSF leaks,
worsening of bowelbladdersexual deficits, epidural and subcutaneous hemorrhages. These are more frequent in patients with spinal cord damage and those with
bilateral DREZ. In patients with incomplete paraplegia, DREZ lesions must not
extend too deeply to avoid additional neurologic deficits. By contrast, in patients
with complete motor and sensory deficits below the lesion, this can be done
extensively on the selected segments. In the series of Falci and colleagues (2002), in
2.3% of patients a temporary pain developed at their new postoperative level of
sensation. A permanent pain (VAS 13) developed in 4.7% of the patients at their
new level of sensation at a follow-up of up to 7 years.
In Nasholds series (Nashold and Pearlstein 1996), long-term relief (pain-free)
of chronic pain from SCIs was obtained in 35% of his patients, with burning
pain and electrical shocks being most responsive. Favorable categories included
patients with incomplete neurological deficit, blunt trauma and conocaudal lesions
with predominant leg pains. Approximately 70% of the paraplegic patients reported
good pain relief immediately after the procedure, although half experienced some
recurrence of the pain postoperatively, usually within the first year. In these patients
the recurrent pain was usually described as less debilitating than the original pain.
Pain in dermatomes at or just below injury (burning, shooting or electrical),
radiating down into the legs and activated by stroking/touching the skin over the
adjacent dermatomes, and unilateral pains usually responded to surgery, but sacrococcygeal and vague diffuse burning pains did not or poorly so (Nashold and Pearlstein
1996). Another favorable group were those who proved to have nerve root avulsions
at operative exposure. Sindou and colleagues (2001) came to similar conclusions.
Radicellotomies performed for pain associated with below-T10 spinal cord lesions are
effective only in patients whose pain has a radiculometameric distribution, i.e.,
the pain corresponding to the level and extent of the spinal cord lesion (end-zone
pain). Pain in the territory below the lesion, especially in the perineosacral area, is not
favorably influenced (while leg pain after caudal lesions is). Nashold also noted that in

299

300

Central Pain Syndrome

18 cases with an intramedullary cyst (syrinx), drainage of the cyst alone did not
suffice, whereas in 18 in whom this was combined with DREZ lesions, 12 good and
2 fair results were achieved.
As mentioned, hyperactive STT and propriospinal cells may contribute to feed the
supraspinal generator (see Chapter 8).
k. S P I N A L R H I Z O T O M I E S : Dorsal rhizotomy is unsuccessful in relieving CP and can
trigger anesthesia dolorosa (Pagni et al. 1993).
l. S Y M P A T H E C T O M I E S A ND S Y M P A T H ET I C B L O C K S : Alajouanine and Brunelli (1935)
and other authors (reviewed in Garcin 1968, but also recently Falci et al. 2002)
thought that sensory afferents may traverse the sympathetic trunk and have a role
in CP. A few patients with BCP and CCP have been temporarily and on occasion
for prolonged periods relieved by sympathetic blockade, whether complete
or not, whether by local anesthetic or guanethidine. This relief appeared to depend
on hyperpathia. When relief occurred, hyperpathia, steady burning and intermittent
shooting spontaneous pain, but not usually deep pain, disappeared. Occasionally,
hyperpathia was relieved but not spontaneous pain, or hyperpathia longer than
steady pain, but spontaneous burning pain was not relieved independently of hyperpathia. However, those studies generally lacked a placebo control and it is not clear
why sympathetic fibers should have a role in CP with allodynia, but not without:
likely, the block reduced sensory barrage tout court, also explaining why not all the
peripheral nerves of the affected region had to have their sympathetic nerve supply
blocked (cases 34 of Loh et al.; see Table 7.1).
The existence of sympathetically maintained pain appears to be more fiction than
scientific fact (Ochoa 1999; Schott 2001) and peripheral blocks, including
sympathetic blocks, are flawed (Bonicalzi and Canavero 1999a, b, 2000). Direct
recordings from human sympathetic nerve fibers have failed to substantiate the
notion of an increased sympathetic outflow in patients with neuropathic disorders
(Blumberg 1988; Janig and Koltzenburg 1991). C-fibers (visceral) are found in
sympathetic nerves, where private lines in the sympathetic nervous system have
been characterized (Hallin and Wiesenfeld-Hallin 1983); occasional positive effects
in CP could be mediated through effects on these fibers (Schott 2001).
The vast majority of authors report no benefit from sympathetic block and/or
sympathectomy in CP (Nashold 1991; Sjolund 1991; Bowsher 1994; Tasker 2001).
The vasomotor, sudomotor and trophic disorders observed in certain cases may just
be reflex phenomena induced by pain (Garcin 1957) secondary to change in mobility.

REPORTS OF SUDDEN DISAPPEARANCE OF CP

Vis sanatrix naturae

There are a few cases of CP which suddenly vanished after long-standing disease.
Nature is teaching us a valuable lesson we must learn from.

Pathophysiology: Human Data

301

Case 1 (Spiegel et al. 1954; Hassler 1970). They observed sudden disappearance
of thalamic hyperpathia due to a lesion of the posterior portion of the thalamus after
a new larger lesion in the posterior ventral nucleus of the thalamus.
Case 2 (Gybels and Sweet 1989, p. 342). These authors treated one patient with pain
in the right leg of 12-year duration after a left cerebral stroke. Several neurosurgical
operations (not specified) had no effect, but morphine (0.05 mg) administered via a
ventricular catheter was followed by a 12 day long complete pain relief and severe
paraparesis; 0.025 mg relieved the pain for 12 hours without motor deficits.
Satisfactory relief continued for 7 months, at which time a major left cerebral infarct
produced a right hemiplegia and complete relief of her pain.
Case 3 (Soria and Fine 1991). Their 62-year-old patient developed an acute
stroke with a right hemisensorimotor syndrome, including pain and temperature
hypoesthesia. Typical CPSP with allodynia developed over 12 months. The threshold
for pain, temperature and light touch was increased, but, when exceeded, the pain
resulting was intolerable. One year following the stroke, a CT revealed a small lacunar
infarct of the left thalamus. Somatosensory evoked potentials revealed absent N18,
N20 and P27 components. Several drugs and other kinds of treatment had no
enduring, satisfactory effect. However, 7 years after the original episode, a second
stroke produced sudden right hemiplegia, motor aphasia and complete disappearance of both the pain and the allodynia. At follow-up, 5 months later, there was pain
and temperature hypoesthesia in the right half of the body. A late CT scan revealed
a well-demarcated, low-density lesion in the left parietal lobe, deep in the centrum
semiovale, adjacent to the body of the lateral
ventricle. Pain was still absent 1 year later
(Figure 7.3).
Case 4 (Hirato et al. 1993). These authors
reported a patient with CP after a putaminal
lesion, in whom many irregular burst discharges
were encountered in the thalamus (Vim-Vc). PET
revealed thalamic hypoactivity and cortical hyperactivity. CP disappeared after a small subcortical
hemorrhage had accidentally occurred near the
cerebral cortex around the central sulcus during
surgery.
Case 5 (Canavero et al. 2001). This woman
developed disabling left hemisoma (C4 sensory
level) CP following surgery for a C45 herniation,
with prominent thermomechanical allodynia in
involved regions. She was refractory to multiple
drug therapy. During MCS, a microdialysis
catheter was inserted into right SI arm area.
Within 48 hours of surgery, the patient started to
complain of a dead flesh sensation to the left
arm distal to the deltoid. A CT scan showed a

Figure 7.3: This patient developed CP after a pure

thalamic stroke (not shown). A further stroke along
the parietothalamic axis abolished the pain.
(Adapted from Soria and Fine [1991], with
permission from the International Association for
the Study of Pain.)

302

Central Pain Syndrome

Figure 7.4: CT scan of a SI stroke selectively abolishing arm cord central pain (left). MR image showing
resolution of the stroke at a time when central pain had relapsed (right). (From Canavero et al. [2001],
with permission.)

right SI infarction and the catheter was removed. For 20 days, the patient complained
of her previous pain, except for the left arm. Thereafter, her CP returned with the
same intensity and characteristics as per before the stroke. During those 20 days there
was complete dense anesthesia of the limb with no sign of allodynia (mechanical and
thermal). Burning pain was absent (VAS/NRS: 0). MRI 8 months later showed
a normal-appearing SI with only a serpiginous area inside (Figure 7.4).
Case 6 (Helmchen et al. 2002). In June 1999 this 58-year-old man experienced
sudden stroke with left-sided sensorimotor symptoms (bar face and neck), with both
lemniscal and spinothalamic deficits. CT showed a hemorrhage in right thalamic Vc.
Three months later, he noticed the gradual onset of a throbbing, burning, aching,
dysesthetic pain on his left side (maximal in the arm) (VAS 8) which became
disabling and was aggravated by movements and cold stimuli. Ten months later,
hemihypesthesia and hypalgesia were unchanged (movement was improved), but
there was mechanical and thermal (cold and warm) allodynia; on CT, a circumscribed hypodense lesion was seen in the posterior right thalamus. MRI also
showed a few subcortical parietal and frontal infarctions in the centrum semiovale
not involving the ACC. Drugs were ineffective. In April 2001, while washing his
hands, he could no more appreciate warm temperature on his right hand, although
being able to differentiate between warm and cold water on his left arm; allodynia on
the left was gone and the spontaneous aching CP on his left side disappeared. He
presented sensory deficits on the right side, particularly severe thermalgesic
hypesthesia. On the left, he could differentiate warm and cold stimuli in his hand,
without a trace of allodynia. There was still hemidysthesia and hypesthesia, particularly in the arm. Simultaneous tactile, but not thermal, stimulation was localized to
the right arm. There was no thermal or algesic sensation in his right hand, while in

Pathophysiology: Human Data

his left it was practically normal. Over the following 2 months, sensory deficits largely
improved on the right side, bar position sense. Concomitantly, spontaneous CP and
to a smaller degree cold thermal allodynia redeveloped on the left side and still
increased over the following months. Almost 1 year later, left CP still persisted, but
without warm thermomechanical allodynia. No CP had yet developed on the right
side. On SSEPs there was prolonged P40 latency on right tibial nerve stimulation.
MRI showed left hemispheric postcentral parietal ischemic infarction (5  4  5 cm)
that involved SI, supramarginal gyrus, SII, external capsule and a very small portion of
the posterior insula, sparing the anterior insula, internal capsule and the left
thalamus.
Case 7 (Daniele et al. 2003). A hypertensive 68-year-old woman developed acute left
hemiparesis with mildmoderate motor impairment, hypoesthesia and tingling
sensation which increased over the days. CT showed a right thalamic hemorrhage.
Several days after discharge, she began to complain of spontaneous pain in her left
limbs, sometimes described as burning and excruciating, and tactile allodynia.
Carbamazepine at 800 mg was only partially effective. Three years later, the pain was
unabated, with partial reduction of hypesthesia. Then, she suddenly developed acute
aphasia. A CT showed a left frontoparietal ischemic lesion plus bilateral lacunar
infarcts. For the next 3 years (until death) her pain and allodynia were completely gone.
Similar cases of disappearance without pathologic confirmation are on record.
For instance, White and Sweet (1955) reported a woman suffering from thalamic
CPSP. Two-staged bilateral orbital gyrectomy gave no relief of pain. However, four
months after operation the pains inexplicably disappeared and the patient was well.
Patient 2 of Michel and colleagues (1990) developed douleur fulgurante en coup de
couteau to the left hand, plus brachial paresis and tactile and pinprick hypesthesia.
CPSP worsened, but 3 weeks later it disappeared with onset of brachiofacial left
hemiplegia, only to be replaced by cheirooral paresthesias; a CT scan showed
a superficial cortical hypodensity straddling right SI/MI. Young and Rinaldi (1997)
state that in one patient, who experienced a right-sided thalamic hemorrhage, neglect
of the left side of the body developed that relieved the patient of her pain, but they
do not state if it was CP. Franzini and colleagues (2003) reported disappearance
of CPSP partially relieved by MCS after an undetailed brainstem stroke.
Finally, there is one single case report (Koszewski et al. 2003) of full CP relief
following thalamoparietal radiation lesioning. Three years previously, a 72-year-old
man developed a right hemispheric stroke. Immediately after the stroke he was
hemiplegic and hemianesthetic. Then sensibility renormalized and his plegia became
a nondisabling hemiparesis. Three months after stroke, he developed burning pain
and allodynia in the left hemibody and became suicidal. In 2002 an MRI showed
a right lesion covering most of the putamen, claustrum, external capsule and part
of the insular cortex; the internal capsule was at least partially damaged.
He was submitted to stereotactic anterior capsulotomy for no clearly explained
reason. During surgery, stimulation of the border between the internal pallidum and
posterior limb of the internal capsule diminished, but not fully abolished the pain.
Two large lesions were done covering the whole border between the posterior limb of
the internal capsule and the lentiform nucleus: in this area only, stimulation controlled

303

304

Central Pain Syndrome

the whole left side of the body. The whole CP

syndrome disappeared immediately after lesioning.
Right after surgery, there was motor worsening
which slowly resolved to previous levels; nociceptive sensibility was fully preserved (implying that a
descending input was interrupted) and no
emotional change was noted. Five months later
the patient was still pain-free (Figure 7.5).
There are many reports of mostly sudden
disappearance of CP after treatment of the
triggering lesion, so called reversible CP.
1. Michelsen (1943) reported four cases of
meningiomas impinging upon the parietal
cortex, in which pain and associated sensory
phenomena in the involved extremities were
present. In his case 4, the pain was completely
relieved by removal of the lesion, and in his case 3
it was relieved for 4 years before it reappeared. His
case 5, with a depressed skull fracture over the
Figure 7.5: Brain MRI scan depicting surgical
anterior and posterior central gyri with cerebral
lesion in the posterior limb of the internal capsule
abolishing central poststroke pain. (From Koszewski contusion, exhibited paraplegia and bilateral
et al. 2003, with permission from VSP, an imprint hyperpathia and hyperesthesia. Position sense
of Brill Academic Publishers.)
was absent in the right leg and diminished in the
left, while pain and touch were recognized and
localized. After debridement, the pain gradually cleared, hyperesthesia receded and
sensation improved.
2. Silver (1957) reported a patient who had a stroke, with hemiplegia and aphasia.
Eight years later, he gradually developed very severe paroxysmal burning pain in the
right arm. An AVM of the left parietal area was diagnosed at angiography and
completely removed in two stages. Under local anesthesia, manipulation, traction
upon and clipping of the component blood vessels reproduced the pain. The pain was
abolished almost immediately and relief maintained for the 5 years of observation.
3. Di Biagio (1959) totally and permanently relieved a CP patient with steady
and intermittent paroxysmal, but no hyperpathia or allodynia, components following
extirpation of a right subcortical parietal tuberculoma.
4. Hamby (1961) reported on a young man who developed severe burning
pain with allodynia in the left upper limb following a car accident. Two years later,
at surgery, the parietal cortex was found to be covered by extensive pools
of subarachnoid fluid. Drainage of these pools revealed yellow, atrophic, leathery
looking cortex resembling that following an old infarct. This area was sharply
separable from normal cortex and extended from the sylvian fissure upward almost
to the interhemispheric fissure, and apparently was limited anteriorly by the
postcentral gyrus. Stimulation over the postcentral gyrus behind the motor points
elicited painful prickling sensations in the upper limb. Stimulation in the

Pathophysiology: Human Data

normal-appearing postcentral gyrus above the arm area elicited painless prickling
sensations in the foot. A transpial incision was made 5 mm deeper than the gutters
of the gyri along the posterior edge of the postcentral gyrus and over three
contiguous parietal gyri. The cortex and adjacent U-fiber areas of the white matter
were easily removed. On the next day the patient had no subjective pain or
dysesthesia or allodynia. The patient remained pain-free 10 years after surgery.
5. Retif and colleagues (1967) reported on a patient (their case 3) who had an
anterior parietal meningioma with purely paroxysmal fit-like pain and a jacksonian
march. Removal was followed by a complete recovery. EEG showed an irritative
pattern.
6. Stoodley and colleagues (1995) reported a 63-year-old woman who gradually
(over many years) developed constant dull pain to the whole right hemisoma (worse
in the face) and an unpleasant tingling sensation on being touched on those areas.
There was no sensory deficit. Neuroradiologically, she harbored a saccular aneurysm
of the bifurcation of the left internal carotid artery extending up to the left thalamus.
There was complete resolution of all her sensory symptoms immediately following
surgical clipping and for a follow-up of 18 months.
7. Potagas and colleagues (1997) described a patient with intermittent pain in
the right arm caused by an otherwise asymptomatic low-grade glioma of the white
matter of the parietal operculum whose pain stopped after excision of the tumor.
8. Fukuhara and colleagues (1999) reported on a woman with a 9-year story of
progressively worsening of episodic deep aching/burning CP to the right hemisoma.
No sensory deficit was present. Neuroimaging disclosed an arterovenous malformation in the corona radiata of the parietal lobe, along the posterior horn of the lateral
ventricle. Embolization achieved complete remission. Transient sensory hypesthesia
was seen (postembolization subparietal ischemia?).
9. Albe-Fessard (personal communication to Barraquer-Bordas et al. 1999) had
a woman with CP in an anesthetic facial area. She had a huge parietal meningioma
with maximal compression on face area. Removal led to CP disappearance and
renormalization of sensibility.
10. Tasker (2001) operated on a patient with a right parietal hemispheral
meningioma presenting with contralateral dysesthetic causalgic pain, which
disappeared after removal of the tumor.
11. We observed several cases ourselves. Pagni and Canavero (1993) reported on
a woman suffering paroxysms of pain, described as burning, lancinating
or electric shock-like, which increased in frequency over the months.
MRI disclosed a posterior T67 meningioma. Extirpation resulted in total remission
over 24 hours, without any further recurrence. Canavero and colleagues (1995)
described a man who developed acute Schneiders syndrome and hyperacute
allodynia to the limbs (worse in the arms in C68 dermatomeres bilaterally) within
30 minutes of a fall. Allodynia was so intense to make sensory examination
impossible. On MRI, there was spondylotic narrowing of the vertebral canal with

305

306

Central Pain Syndrome

large osteophytes at C47, particularly on the posterior aspect. A voluminous spur

jutted out of the right posterior aspect at C7; the C5/6 disk was posteriorly excluded,
impinging upon and nicking the anterior surface of the dural sac, with greatest
narrowing at C46. Upon reawakening from surgery (C5/6 discectomy plus
stabilization), the allodynia had completely disappeared. Sensory examination at
this time showed thermoalgesic hypesthesia in the four limbs. Two weeks later,
typical CCP involving the four limbs appeared and gradually worsened. Pagni and
Canavero (1995) relieved CP involving one leg after aspiration of a benign
intramedullary cyst (follow-up 10 years; unpublished observations). Canavero
(1996) reported on a woman who developed burning pain in her left arm and,
episodically, in the whole hemibody due to a bleeding cavernoma in the white
matter deep to the inferior parietal lobe. CP totally regressed after the bleeding
cleared, only to return with a new bleeding years later (unpublished observations).
Canavero and Bonicalzi (2001) reported on two patients. The first suffered from
severe burning pain and allodynia to one leg which totally vanished within 24 hours
of extirpation of a cystic tumor at conus level (follow-up: 3 months). The second was
immediately relieved of her intermittent CP following embolization of an aneurysm
at the vertebralPICA junction impinging on the medulla (follow-up: 2 months).
Finally, we relieved a 54-year-old woman of her pain to the left leg, misdiagnosed
as sciatic pain, after shunting a large parietooccipital arachnoidal cyst. Another
woman with a meningioma compressing SI had painful fits to the hemibody,
abolished after surgery (Canavero and Bonicalzi, unpublished observations).
CP associated with MS may often present during acute relapses and spontaneously
vanish as the relapse clears (e.g., Portenoy et al. 1988). Some cases of CP receded after
shunting for syringomyelia (e.g., Suzuki et al. 1985; Milhorat et al. 1996; Attal et al.
2004; see also Chapter 8) and it is reported that type I Chiari malformationassociated neurogenic pain (but not particularly sensory loss) responds well to
surgery (Meadows et al. 2001; Bejjani and Cockerham 2001).
However, generally speaking, CP is a chronic pain, which usually stays with
patients for the rest of their lives. On rare occasions, it may gradually subside even
after prolonged periods (CPSP; Greenspan et al. 1997; Kim 1999). According to
Schott (2001), CP can disappear spontaneously even after many years, temporarily or
permanently, generally slowly, but he does not back up this assertion with personal or
published evidence; however, he had a patient with unremitting CPSP for 15 years
except for 8 hours of 100% relief during a flight (similar to cases of causalgia and
Parkinsons disease). Slow disappearances would feature ever longer pain-free
intervals, although, when present, pain would be as severe as ever. Andersen and
colleagues (1995) reported that in two patients CPSP disappeared spontaneously:
one had evoked dysethesia and shoulder pain at 1 month and another, with a lower
brainstem infarction, complained of ocular pain with a Horner syndrome. The CPSP
case 1 of Michel and colleagues (1990) simply reported an abatement of his pain.
Garcin (1968) stated that regression of brainstem CP is exceptional, but a few cases
were seen.

PIECING TOGETHER THE EVIDENCE

. . . in the light of knowledge finally achieved, deductions seem almost obvious and can
be understood by any intelligent student; but the experience of research, gropingly in the
darkness, with its profound anxiety to succeed and its alternating character between certainty
and discouragement, can only be understood by him who has experienced it. (A. Einstein,
1935)

The evidence reviewed strongly suggests that CP may be understood as the result
of a localized reverberation loop between the parietal cortex (SI, and perhaps SII)
and the sensory thalamus (Vc, core and shell) with a supporting role of Vim, CL
and its SI projections, and pulvinar (Canavero et al. 1993; Canavero 1994), as this is
the only mechanism able to explain pain disappearance following lesions limited
to the subcortical white matter (see Box 8.1). This dipole is exquisitely adjusted
to explain somatotopographical pain distribution in CP (Canavero 1994). The loop,
with its descending excitatory arm, is engaged bilaterally, with contralateral
or,
in some cases, ipsilateral
predominance. In those rare cases with complete SI
or thalamic destruction (e.g., maxithalamotomies), the reverberant loop can be
activated contralaterally. CP appears to be more frequent after right-sided lesions,
perhaps due to lateralization of norepinephrine. Since the evidence points to
a major role of this arm in CP sustenance (see also Yamashiro et al. 1991), we
propose that STT lesions (or simple interference, without actual sensory loss)
unbalance the normal oscillatory corticothalamic dialogue, starting in SI, where
GABA levels drop acutely, and induce changes caudad along a diffuse spinotruncothalamic reticular core (see in Gybels and Sweet 1989; Nandi et al. 2004), which
becomes hyperactive. The end result is bilateral facilitation from multiple top-down
(locked SI) and bottom-up (cord and brainstem reticular) sources. Intrathalamic
activity hinging on TRN can be entrained by corticothalamic oscillations and drive,
in turn, the cortex.
The substratum of release is speculated to be a genetically defective GABA A
receptor at cortical sensory level, which may be present in about one-fifth to
one-third of the population (see the speculation of Zimmermann [1991] of
a molecular defect at the level of inhibitory synapses as the underpinning of
neurogenic pain).
307

308

Central Pain Syndrome

Box 8.1 Historical note

Livingston (1943) proposed that chronic pains following peripheral neural injury were the result of a vicious
cycle set up as a central perturbation of function in the internuncial neuron centers of the spinal grey
matter by an irritant focus. He wrote:
once the central process is started it assumes the major role . . . If the trigger point is removed early,
the process may subside spontaneously. If the process is permitted to continue . . . even a removal
of the original irritant may not be sufficient to establish a cure . . . the central disturbance is the essential
factor in many diseases, and that there should be better means for eliminating pain than by a chordotomy or
posterior root section or other anatomic interruptions of nerve continuity.
Building on the work of Lorente de No in the 1920s on the concept of closed self-reexciting chains within
neuronal pools, Von Hagen (1957) wrote:
chronic pain syndromes, i.e., sustained pain after the disappearance of the original impetus, cannot
develop unless cortical components are involved . . . [it] is the product of reverberating circuits in
the nervous system of which the cortical components are of great importance . . . the various emotional
reactions related to this state, namely, preoccupation with symptoms, introspection concerning among
other things the memories of the pain and anticipation of the future, depression, and anxiety, as well
as emotional stress brought on by other conflicts, also serve to reinforce the reverberating circuits
and further perpetuate the disability.
Likewise, Talairach and colleagues (1960) believed that:
Certain factors favorise the concepts of the cerebral cortex taking part in the elaboration of the
painful sensations, the parietal lobe being directly involved . . . it seems proved that there are reverberating
thalamo-cortico-thalamic circuits, capable to modify at any time the modalities of the afferent impulsions
and their non specific incidences.
Thus, they coagulated the white matter beneath the parietal lobe in four causalgic, two phantom pain and
two facial pain patients, as
a sub-cortical lesion localised at the cross-road of thalamo-cortical paths should achieve a sufficiently
generalized section of the painful afferences, and disturb in the same time this modulating system
of thalamic activity or more generally the regulation of the areothalamic couple.
In particular, they wanted to interrupt the path to SII. Sano and coworkers (1966) hypothesized that CP due
to thalamic lesions might be due to a reverberating circuit between hyperirritable cells of the lateral and medial
thalamic nuclei. Emmers (1981) speculated that peripheral nociceptive sensory input may impinge upon
a preexisting low-threshold thalamic discharge system to activate a self-sustained reverberating system of parallel
facilitatory feedback loops. The reverberating system between the lamellar, CM-Pf and SII neurons would sustain
a more or less constant level of centrally activated pain referred to a given body part, depending upon the somatic
representations of the involved SII neurons.

The reticular formation (RF) becomes bilaterally primed right after injury;
one possible role, besides feeding the loop, could be to engage a dedicated paincoded sensory loop contralaterally (this may also occur by corpus callosum-mediated
transfer of anomalous oscillatory activity to the opposite side). Since the switch
from unconscious to conscious state likely correlates with the strength of activation
in a given area (Zeki and Ffychte 1998), a hyperactive RF feeding on the corticothalamic loop may contribute to the conscious feeling of CP. Thus, in patients
with deafferentation pain the medial midbrain tegmentum becomes hypersensitive
to stimulation, and that along with posterior thalamus, thalamic radiations and

Piecing Together the Evidence

somatosensory cortex, acquires the property, absent in somatic pain syndromes,

of generating not only a painful conscious awareness but also a reasonably
accurate reproduction of the patients pain, but only in the already painful sites;
due to deafferentation, mesencephalic reticulo-thalamic-cortical circuits become
sensitive not only to electrical stimulation but also to natural neural input (Tasker
et al. 1980).
That the cortex (above all SI) plays a leading role is supported by this being
the initial target of IV propofol and IV ketamine, which effectively relieve CP.
Lack of opioid efficacy (and CS superiority on other neuromodulatory techniques)
in CP may be due to this highly corticalized mechanism, notably to a dearth of
opioid receptors in SI. Even in some studies of acute pain, cortical activity can be
detected before subcortical responses appear (Casey et al. 2001). Rosso and colleagues
(2003) recorded SEPs 2 hours before and 3 after percutaneous cordotomy in 7 cancer
patients and found that nociceptive STT denervation may induce a rapid modulation
of cortical (SI), but not spinal or brainstem, neuronal activity along the lemniscal
pathway. On the other hand, CP usually requires an at least partially intact thalamus,
ipsi- or contralaterally, as proved by too massive a thalamic destruction being
incompatible with CP (see SEPs data and Spiegels case of remission in Chapter 7).
Ohye (1998) found that, in CP, the initial hemorrhage or infarction in the
thalamus is rather small (less than 1 cm in diameter), in cases that developed CP
within 1 year; patients with massive thalamic involvement following initial stroke
did not manifest CP, but only hypesthesia in general.
Absence of pain is a homeostatic, dynamic condition between pro- and antinociceptive CNS activities. Malfunction (even at a molecular level) of inhibition
(including transformation of inhibitory synapses into excitatory ones), possibly on
a genetic basis, with changes in dynamic cortical network strength, may allow
pathologic sensory deprivation or alteration of ascending sensory or descending
modulatory fibers to disrupt the normal, homeostatic pattern of neuronal activity
of sensory systems. This alters the network functional mode to the point that
a dedicated oscillatory (resonant) pain loop (perhaps seen as coherent theta
activity) is switched on in the thalamocortical axis responsible for subjective states
indefinitely (Canavero 1994), with other brain areas playing a corollary role.
In other words, the same mechanism responsible for the genesis of consciousness
can generate CP when its organization and timing are altered by disrupted inhibitory
dynamics.
Different qualities of pain, but also different neurometabolic findings, may
be explained by individual degrees of activation of the same cells or activation
(frequency discharge/oscillatory changes) of several sets of cells, in different cortical
layers and thalamic nuclei, depending on site and extent of damage. The loop
would be under the influence of cognitive, emotional and attentional networks,
explaining fluctuations in time of CP. This mechanism would apply to all CP
conditions, from dorsal horn to cortex.
One consequence of the establishment of such loop would be the functional
dissolution of processing circuits, with loss of information, triggered by stable
neural anomalies that hinder correct data estimation. The thalamocortical system
becomes less flexible (efficient) in sampling inputs and evaluating information

309

310

Central Pain Syndrome

both from evoked and spontaneous sensory stimuli, flexibility implying the capacity to occupy different bands of discharge frequencies. In the cortex, stimuli
are coded with a loosened information exchange and peculiar sparse clusters of
connectivity (Biella et al. 1999), i.e., information processing decorrelates. Loss/
distortion of proper spatiotemporal sequence/somatotopy of incoming impulses
at cord, brainstem or thalamocortical levels in CP has already been entertained
by past authors (e.g., Foix et al. 1922; Zuelch and Schmid 1953; Donovan and
colleagues 1982).
Thus, we may now have a cure for CP, i.e., a stereotactic lesion in the subparietal
white matter, in some cases bilateral, targeting the descending facilitatory arm
of the loop. Neurosurgical experience shows that, once the sensory component of
chronic pain is abolished, pain affect also is renormalized (not vice versa) and
this would be the case for the proposed intervention. Of course, this intervention,
carrying the same morbidity/mortality of, for example, deep brain stimulation,
should be reserved to patients refractory to therapies detailed in Chapters 5 and 6.
The proposed scenario also would explain successes with electroconvulsive therapy
(Canavero 1994; Chapter 6).
Importantly, the present framework nixes the idea that in chronic pain, the
widespread nature of pain (matrix) processing precludes effective focal treatment
by neurosurgical means (Melzack 1991): CP can be reversed (Chapter 7, Section 4).
Also, the vast majority of proposed theories, including exclusively based thalamic
theories of CP, collapse on such clear-cut observations.

EXPLORING THE THEORETICAL FRAMEWORK

Classic neurophysiology has focused on the encoding of information through
changes in the firing rate of neurons, but measuring mean discharge rates may
not yield the full information transmitted. When networks of neurons interact,
the result is often rhythmic activity within defined frequency ranges that can
engage in temporal synchronization and de-synchronization ((de)correlation),
i.e., changes in the bands of oscillations convey additional information to
neuronal firing rates. Neurons fall into step with one another forming ensembles
firing in relative synchrony for brief periods, before some neurons drop out of
synchronization to join another ensemble. Synchrony between trains of action
potentials has both oscillatory and non-oscillatory components (Jones 2001).
Importantly, they cannot be detected as CBF changes, since sensory discrimination,
for one, may require a limited fraction of neuronal population and a change in
synchrony may suffice. These assemblies are dynamic and shifting and are associated
with perception.
According to Llina`s and Pare` (1991, 1997): only a minor part of its connectivity
is devoted to the transfer of direct sensory input. Rather, most of the connectivity
is geared to the generation of internal functional modes, which may, in principle,
operate in the presence or absence of sensory activation [p. 521]. . . the number of
cortical fibers projecting to the specific thalamic nuclei is larger than the number
of fibers conveying the sensory information to the thalamus. Thus, a large part

Piecing Together the Evidence

of the thalamocortical connectivity is devoted to re-entrant or to reverberating

activity . . . the insertion of neurons with intrinsic oscillatory capabilities into this
complex synaptic network allows the brain to generate global oscillatory states
[i.e., population coding] which shape the computational events evoked by sensory
stimuli (p. 526).
Oscillations are generated in different sectors of the thalamus or cerebral cortex,
even if they are disconnected from other structures; in the intact brain, these
coalesce within complex wave sequences owing to neuronal interactions. Large
ensembles of cortical and thalamic neurons discharge synchronously at stereotyped
frequencies associated with different conscious states. During alert wakefulness,
high-frequency oscillations occur spontaneously or as part of sensory-elicited events
in the relay nuclei of the thalamus and the cortical areas to which they project,
binding distributed aspects of sensory perception (consciousness). Thus, disruption of oscillation and/or temporal synchronization may be a fundamental
mechanism of neurological disease (Farmer 2002). Stochastic oscillating or clustered
discharges with the same mean discharge rate may have very different effects on
transmitter release, temporal summation of postsynaptic potentials, long-term
changes of synaptic strength and second messenger effects (Sandkuehler 1996; but
see critique in Pareti and De Palma 2004) (Box 8.2).
CP can be understood inside this framework: CNS lesions are not simply
depriving the brain or parts thereof of afferent input; they are disrupting an ongoing
pattern of neuronal activity.

TRIGGERING PERSISTENT OSCILLATION

Three approaches appear promising in the present context and these will be briefly
discussed. The gist is bistability, i.e., the property of switching between two stable
states (e.g., pain
non-pain).
1. Neural networks. Neural networks are dynamically regulated entities, constrained
by their anatomical connectivity and membrane properties of the component
neurons. They can be tuned and configured into several operational modes,
each depending upon the expression and modulation of the constituent cellular,
synaptic and network building blocks, and in accordance with the conditions of the
moment. By changing the properties of selected synapses, cells or pathways, the operation
of a network can be dramatically altered, e.g., from mutual inhibition to mutual
excitation; controlling inputs may turn an oscillatory circuit on and off, and the
functional connectivity can be reconfigured by ascending and descending CNS
influences, i.e., a radical rewiring. Control systems able to switch the operational
mode of nociceptive cells exist in the CNS (Willis 1985). Also, a single network can
participate or generate a large repertoire of outputs, which may be determined,
among others, by sectorial damage to targets of innervation (e.g., Selverstone and
Moulins 1985; Getting 1989). Persistent activity can arise from a large neural network
that involves recurrent excitatory loops through reciprocal excitation between the

311

312

Central Pain Syndrome

Box 8.2 Thalamocortical rhythmicity

The neocortex and thalamus are a unified oscillatory (reverberant) machine and they work in concert (Jones
2001). The main cortical projections from Vc end in layers IV and III forming dense, topographically organized
arbors that synapse mainly with dendritic spines, plus some branches ending in layer VI. Synapses formed
by Vc TC axons in layer IV are less than 10%. Possibly, individual TC synapses produce a stronger synaptic
drive than intracortical synapses. CT fibers go back to correspondent TC neurons only to a limited extent.
Anatomically and functionally, feedback connections are quite different from feedforward connections: although
much more numerous than the latter between the same structures (Ullman 1995; Sillito et al. 1994), individual
feedback projections mediate less powerful effects, having more sparsely branching, widespread arbors, with
fewer less effective synapses largely on distal dendrites (versus proximal dendrites of feedforward fibers), and
also being less precisely focused within a sensory representation, reaching a larger proportion thereof
(see references in Kaas 1999; Ergenzinger et al. 1998). Overall, top-down feedback connections seem better
designed to stimulate weakly larger groups of neurons and modulate ongoing activity (versus feedforward
connections creating activity in smaller groups of neurons). Following, for example, NMDA block of corticothalamic
(CT) SI cells, these fibers might focus transmission (small RFs), via GABA interneurons in Vc and TRN
(and eventually other thalamic nuclei), thus placing thalamic plasticity under cortical SI control (Kaas 1999).
Also, top-down influences may alter the overall functional nature of SI and layer-specific mechanisms of sensory
processing (Krupa et al. 2004). On the other hand, intranuclear inhibition through the TRN may affect RFs
of cortical neurons.
Thalamocortical rhythmicity is driven by the thalamic GABAergic reticular nucleus (TRN) that projects to
almost all thalamic relay nuclei (Shepherd 2004). It receives excitatory glutamatergic inputs from axon collaterals
of thalamocortical (TC) fibers that traverse it on their way from thalamic relay nuclei (including Vc) to cortex
and of corticothalamic glutamatergic fibers that project back from cortical layer VI to thalamic relay nuclei.
However, it is possible that not all TC/CT axons give off collaterals to TRN. TRN cells are highly interconnected
through inhibitory, mainly GABA A, dendrodendritic or axodendritic synapses: they can generate rhythmic
sequences of LT Ca2 spikes (through an interaction between the LT Ca2 current and a Ca2 activated K
current), which, through activation of GABA A receptors, generate similar bursting in TC cells. These in turn excite
TRN cells, closing a disynaptic loop. TRN cells show the highest levels of tonic activity during heightened vigilance.
Some GABA TRN cells also project to local inhibitory neurons located in different dorsal thalamic nuclei.
Thus, some TC cells may be disinhibited inside a surrounding core of inhibition. In primates, the TRN can be
divided into a number of sectors each concerned with a different function. Each sector is connected to more
than one thalamic nucleus and to more than one cortical area; each sector has topographically mapped
connections with the thalamus and cortex. For instance, Vc relates to one sector of TRN, bidirectionally, and
TRN also receives from SI. Connections are not the same for each sector: TRN acts as a nexus where several
functionally related cortical areas and thalamic nuclei interact modifying TC transmission through the inhibitory
connections that go from TRN cells to TC relay cells. In the somatosensory system, both first- and higher-order
nuclei project to the same sector (e.g., Pom
higher order
relays to SII; Vc
first order
to SI). Although
complete human data are not available (Guillery et al. 1998), humans appear to have a network of intrinsic
thalamic and cortical GABA interneurons. A mutual inhibitory coupling exists between TRN cells, responsible
for nuclear oscillation, synchronized by CT input. TRN cells may also project to contralateral dorsal thalamus in
the intrathalamic commissure, potentially influencing the cerebral cortex and basal ganglia of both hemispheres
(Steriade et al. 1997).
The cortex has a powerful role in controlling the coherence of thalamic oscillations. CT synaptic volleys
succeed in synchronizing pools of thalamic cells by activating GABA TRN cells that project to thalamic relay
cells and hyperpolarize them. Slow cortical oscillations initiated apparently in layer 5 as an excitatory interaction
between pyramidal neurons propagate through the neocortex. This generates a depolarized state through recurrent
excitation regulated by inhibitory networks, thus allowing local cortical circuits to enter into temporarily activated
and self-maintained excitatory states. At each step of the pathways that link various neocortical areas, CT neurons
impinge on TRN cells that in turn produce inhibitory rebound sequences in dorsal thalamic relay neurons projecting
in the reentrant corticopetal systems, thus changing the time course and synchronization of intracortical events.
Short-term plasticity processes, i.e., persistent and progressive increases in depolarizing synaptic responses and

Piecing Together the Evidence

313

BOX 8.2 Thalamocortical rhythmicity (continued)

decreases in inhibitory responses, can lead to self-sustained oscillations owing to resonant activities in closed
loops. The repeated circulation of impulses in reverberating circuits could lead to synaptic modifications in target
structures (Steriade 1999; Sanchez-Vives and McCormick 2000). Under certain physiological conditions one
neuronal neocortical electrophysiological type can be transformed into another by small changes in membrane
potential or synaptic activities inside thalamocorticothalamic loops (Steriade 1999; Sanchez-Vives and
McCormick 2000).
Neurons with intrinsic oscillatory properties (cell-drive oscillators), networks of non-intrinsically oscillating
GABA interneurons (network oscillators based on reciprocal inhibition, recurrent cyclic inhibition, but also on
recurrent excitation), driven by tonic metabotropic glutamatergic input or (more often) mixed oscillators and
long loop thalamocortical interactions all contribute to both the occurrence of oscillatory activity and their
frequencies. Longer range synchrony in the neocortex could occur by resonating with the thalamocortical
loop (Jefferys et al. 1996). Recurrent processing may have a specific role for perceptual awareness (Supe`r et al.
2001).
Individual neurons can have frequency preferences that enable them to generate spontaneous oscillations
or respond best to inputs within a narrow frequency window (low- and/or high-pass filtering behavior
creating a notch filter leading to resonance), with a role in determining the dynamics of coherent brain activity.
Resonance and spontaneous oscillations can coexist in the same system, being two aspects of the same basic
phenomenon of frequency preference. A resonant system evolves continuously into a spontaneously oscillatory
system as the amplifying conductance is increased. The frequency of the oscillations of the resonance is set by
the properties of the resonant conductance. There are three classes of frequency-dependent mechanisms
in central neurons: solitary resonances, resonances arising from interaction with amplifying mechanisms
(e.g., NMDA mediated) and spontaneous oscillations caused when a resonant current interacts so strongly with
an amplifying current that the resting membrane potential becomes destabilized. In other words, slowly activating
currents that actively oppose changes in membrane voltage produce resonance: the frequency of resonance
is voltage-dependent. The scope of resonance may be to help integrate inputs to neurons (Hutcheon and
Yarom 2000).
The overall setpoint of the thalamocortical system is modulated by several inputs from brainstem, hypothalamus
(activating) and cortex (layer 6) (Shepherd 2004). Thus, there is a high concentration of 5HT/histamine
input in CL and related nuclei, while TRN cells are excited by NE and 5HT (inhibiting TC output) and inhibited
by Ach (M2) from Meynerts nucleus (during novelty or danger) and GABA (e.g., from basal ganglia or other
TRN or inhibitory interneurons) (facilitating TC output): the process can be highly selective, creating foci of
inhibition or disinhibition, e.g., in Vc. The GABAergic projection from basal forebrain may target TRN, but not Vc.
Transition from burst to tonic mode in TC cells results from 5HT, NE, Ach, histamine, nitric oxide and glutamate
input, vice versa only from glutamate input. NE/DA fibers modulate the loop, by acting on layers 5
(thalamoreceptive) and 1 (where dendrodendritic synapses between TC projections from CL and those from
bursting pyramidal cells in layer 5 exist).

cortex and thalamus (Wang 2001), but can also be produced locally within a cortical
area from reverberatory excitation, stimulus selectivity being formed by recurrent
inhibition within a columnar cortical network (Goldman-Rakic 1995). Persistent
activity can also be maintained by reciprocal loops between cortical areas (see
references in Wang 2001); extensive horizontal excitatory connections are known to
exist, especially in layer 2
3. Feedback excitation can also originate from regenerative
membrane dynamics of single neurons: voltage and Ca2 gated ion channels could in
principle generate bistability between a resting and an active state, sustained by a
plateau potential. Activation of relevant ion currents could require neuromodulatory
signals such as acetylcholine. Also, the coupling strength of the neural network of the
brain changes periodically, with a cyclic alteration from a central to a parallel

314

Central Pain Syndrome

processing mode of information, reflecting state transitions from synchronized, low

complex EEG activity to desynchronized high complex activity and vice versa, with a
disturbance of temporal order (Tirsch et al. 2004).
2. Nonlinear dynamics. The emergence of patterns in open, non-equilibrium
systems (e.g., the brain) is governed by their stability in response to small
disturbances and predicts macroscopic transitions between patterns of differing
stability (Meyer-Lindenberg et al. 2002). Using ergodic nonlinear dynamics,
discharge patterns can be represented by an attractor (Stewart 1997). The emergence
of a persistent attractor state (attractors with non-integer dimensions are called
fractals) requires that excitatory connections in a recurrent network are sufficiently
strong; when the strength of excitatory connections between neurons within each
subpopulation is increased beyond a critical threshold, persistent activity appears as an
all-or-none phenomenon. Below the critical threshold, only the spontaneous state
exists; above it, the spontaneous activity state is still dynamically stable to small
perturbations, because at low firing rates excitation is effectively counteracted by
feedback inhibition. However, if a stimulus generates a transient high activity in a
neural subpopulation, recurrent reverberation is now sufficiently powerful to drive
this group of cells to escape from the spontaneous state. A higher firing activity
leads to an even larger recurrent synaptic excitation, which becomes sufficient to
sustain a persistent active state after the stimulus is withdrawn. The firing rate is
eventually stabilized by negative feedback. As a result, a stable attractor of persistent
activity with an elevated firing rate is realized, which coexists with the stable
spontaneous state, i.e., chaos can synchronize. Among possible contributors to
control of firing rates are outward ion currents in the cell, feedback inhibition, shortterm synaptic depression and saturation of the synaptic drive at high frequencies. A
prediction from attractor models is that persistent activity depends on the strength of
recurrent excitation in an abrupt manner, so that activity could disappear suddenly
when excitatory synaptic transmission is gradually reduced by pharmacological means.
Through a complex mechanism involving temporal filtering of rhythmic signals
through resonance, subthreshold oscillations and bursting (Izhikevich et al. 2003),
the brain can reorganize itself dynamically within a few milliseconds, without changing
the synaptic hardware. Structured excitatory connectivity can arise from a columnar
organization or through hebbian long-term plasticity (but synaptic reverberation
is also possible). Persistent activity can be stored in the form of a bump attractor,
a spatially localized persistent activity pattern naturally arising from a network
connectivity under certain conditions and sustained by recurrent synaptic excitation
within a local group of pyramidal cells. Stable bump attractors typically require
that lateral inhibition is spatially more widespread than excitation with interneurons
with broader tuning curves and/or projecting widely to their targets. A loop is more
stable if the networks recurrent synapses are primarily mediated by NMDA
receptors.
3. Small world networking. The brain has a mixture of order and chaos,
with local thick connections and more random global connections. This paves the
way to small world networks (Strogatz 2003): regardless of its size, any two points
within neurons are always linked by only a small number of steps. When 10
20% of

Piecing Together the Evidence

neurons participate in shortcuts, the network forms self-sustaining loops of activity.

For instance, following an activating pulse, region A may activate region B through a
shortcut that would similarly trigger C; a shortcut from C sparks A again, completing
the loop. A second strong activating pulse may shut the whole system down again.
This conceptual approach directly leads to bistability (Roxin et al. 2004). In this
scenario, shortcuts are fast relay channels that allow reciprocal influences to spread
rapidly in the entire population. A small world architecture entrains a more efficient
global coordination (Buchanan 2002; Strogatz 2003).

WHAT RELEASES CP?

While a variety of cultural, psychological and physiological factors contribute to
variability in both clinical and experimental contexts, the role of genetic factors
in human pain sensitivity is increasingly recognized as an important element,
notably genetic predisposition acting, for example, at receptor level (Mogil 1999;
Kim et al. 2004). Women, despite possessing an adjuvant nicotinic spinal antinociceptive path locally mediated by estrogens, are known to be more responsive
to noxious stimuli (Woodrow et al. 1972) and to be at much greater risk for
developing a large number of pain syndromes. Pain thresholds increase with age
(Schludermann and Zubeck 1962), due to an apparent age-related change in the
central primary afferent response to peripheral insult (Friedman 1991). PET studies
reveal that humans widely differ in baseline and pain-induced levels of endogenous
opioids (Zubieta et al. 1999, 2001), with larger activation of the opioid system
correlating with lower sensory and affective ratings to the sustained pain stimulus.
Pain sensitive persons show more frequent and more robust pain-induced activation
of SI, ACC and PFC than stoic ones (Coghill et al. 2003). We proposed that there
may be a genetically determined oscillation threshold of pain-coded thalamocortical neurons, which may be particularly low in CP patients (Canavero 1994).
If we consider single diseases originating CP, it is difficult to say whether CP is
truly more frequent in one or another, due to a lack of epidemiological evidence.
For instance, there may be a difference among compressive (e.g., meningiomas)
versus disruptive (e.g., ischemia) lesions. On the other hand, purported rarity of
tumor-associated CP may either depend on underdiagnosis, infrequence of parietal
lobe lesions compared to all possible brain sites or be related to the fact that many
tumors displace, rather than destroy, neural tissue during the early stages of their
development; however, compression can be enough to trigger CP. What is now clear
is that the incidence of CP at thalamic, brainstem and cord (too few epidemiological
data are available for the cortex) differs little (Chapter 1): after thalamic stroke with
sensory symptoms, mesencephalic stroke, spinal cord injury (below-level pains only)
and MS, this runs at, respectively, 17
18%, 15
25%, about 25% and about 18%.
It might seem that STT damage leads to CP in a similar percentage of patients,
regardless of level. This would nix older theories that differential incidences of
CP are due to anatomical proximity rather than not of different pain bundles
(STT versus SRT) at cord, brainstem and thalamic levels (e.g., Cassinari and Pagni
1969; Nathan and Smith 1984), an anatomically moot observation. This population

315

316

Central Pain Syndrome

of patients must have something in common besides anatomical damage

(idiosyncrasy), since STT damage alone is necessary, but not sufficient to release CP.
During the past 100 years or so, several theories have been proposed to justify CP
release. Following the original proposal of Head and Holmes (1911) to explain the
thalamic syndrome (exaggerated responses in cases where the thalamic centre has
been freed from control of descending cortical origin), imbalance/disinhibition
(escape) theories prevailed, in light of their explicative potential (Botterell et al.
1954).
1. STT inhibits ML. According to a theory (Beric 1993, 1999, and references therein),
stroke or SCI patients with complete sensory ascending dysfunction (STT, perhaps
also SRT, and DC/ML) do not
bar rare exceptions
develop CP (however, this
does not mean that 80% of sensory stroke patients who do not have CP have
complete destruction of sensory pathways, which is not the case). Instead,
dysesthesias are reported in incomplete SCI patients, all with mild, moderate or
severe disruption of STT modalities and partial or complete preservation of A-beta
mediated (DC/ML) modalities, with movement, gait and even, in some instances,
SCS (e.g., in one ASAS case; see also Cole et al. 1987) exaggerating the dysesthesias.
TENS, SCS, thalamocapsular DBS and even CS may worsen CP in some patients (see
Chapter 6). Triggs and Beric (1994) found that 6 (1 thalamic lesion, 2 a brainstem
lesion and 3 a cortico-subcortical lesion) of 48 stroke patients had functionally
limiting dysesthesias induced in the setting of dynamic mechanical allodynia or
by neuromuscular electrical stimulation (NMS). All these had relatively preserved
sensibilities attributable to DC/ML function. Even during remission, dysesthesias and
pain could be triggered by additional afferent input to the DC/ML system; gentle
touching of partially deafferented dermatomes evoked dysesthesias in two of their
anterior spinal artery syndrome (ASAS) CP cases. Since dysesthesias usually appear at
the time when dorsal column modalities of sensations can again be elicited, in the
face of severe and still complete interruption of STT functions, the recovery and
activity in the DC/ML system may be surmised to set off a chain of events at
thalamocortical levels. In fact, CP often arises as sensory (and motor) loss improves
(Schott 2001a). One exemplificative patient (case 22, Mauguie`re and Desmedt 1988)
showed an aggravation of the pain, as lemniscal transmission improved and cortical
SEPs partially recovered. None of 4 patients with combined lateral and medial
medullary stroke developed CP (Kim et al. 1995b), nor did ASAS patient 3 of Triggs
and Beric (1992), who had the most severe anterolateral system dysfunction. These
authors (Triggs and Beric 1993) also reported disinhibition of somatosensory evoked
potentials in a patient with ASAS. On the other hand, cordectomies which completely
destroy all ascending afferences should quench below-level CP and do not. SCS
actually relieves, at least initially, a minority of incomplete SCI CP patients and so
does TENS (see Chapter 6). CP occurs in cases with complete spinal transection or
following supratentorial lesions affecting both STT and ML sensibilities. Benefit is
also seen after electrophysiology guided DREZ lesions for both at- and below-level
pains. Cordotomy can relieve evoked pains (without worsening spontaneous pains)
in several patients. In light of these objections, Beric concluded that the suggested

Piecing Together the Evidence

mechanism is only responsible for dysesthesias and not pain. Actually, lemniscal
fibers do play a role in tactile allodynia (see below).
2. ML inhibits STT. Fabritius (1907) believed that it was damage in the posterolateral
columns of white matter of a corticofugal pathway that was correlated with
the appearance of the spontaneous pains that beset some of his patients after
SCI. Forster (1927, 1936) observed that, after division of the posterior columns or the
medial lemniscus, normally painful stimulation becomes excessively painful,
stimuli normally painless become painful and, above all, spontaneous unpleasant
dysesthesias and more or less severe spontaneous pains in those parts of the
body corresponding to the lesion in the posterior column may occur. After weeks to
months, symptoms gradually disappeared, but in some cases persisted for years.
Forster (1927) wrote: in the area of the posterolateral columns, possibly in the
boundary zone with the gray matter, a corticofugal pathway runs which exerts a
damping influence on the pain system associated with neurons of the posterior horn
whose loss leads to increased excitability of these posterior horns (see also Riddoch
1938; Frazier et al. 1937). Pool (1946) noted in his posterior cordotomies that
The application of cool or warm objects to the skin produced an exaggerated and
disagreeable sensation of cold and warm respectively, causing the patient to flinch.
Orthner and Roeder (1966) believed that CP resulted from lesions of the ML (but not
of the SRT) tract. The patients reported by Nathan and colleagues (1986) with
posterior column lesions
had an abnormally increased sensation with pricking, warm and cold stimuli,
with rubbing the skin and with any stimuli that caused tickle. Pain on pressure had
a lower threshold than in the normal; and painful stimuli felt more unpleasant and more
painful than in normally innervated regions. In fact, all forms of sensibility relying on
impulses in the spinothalamic complex were increased. When the lesion had been made
in one posterior column, these effects were ipsilateral to the lesion and occurred in
the segments deprived of posterior column fibers. That this increased sensation is
accompanied by an increased discharge in the relevant neurons is suggested by the fact
that in one patient the threshold of the flexor response to noxious, warm and cold
stimuli was lowered.

Sweet (1991) believed that it seems possible that the pain following posterior
column lesions . . . is due to elimination of tonic suppressor impulses in this same
region. One of his patients with bilateral cancer pain in the torso had a bilateral
cordotomy in two stages at high dorsal levels. Both incisions accidentally generously
incised both posterolateral columns. Months after the second operation, CP
developed first on one side, then on the other, in analgesic areas. The only type of
posterior column sensory deficit was a loss of vibration sense in the right leg. Autopsy
revealed extensive demyelination of the ventral 70% of the posteromedial columns of
white matter and less intense degeneration of the ventral part of the posterolateral
columns of white matter. A second patient developed CP and spasticity in all four
limbs since birth, from a central cerebral lesion. He underwent a two-stage bilateral
upper thoracic cordotomy, which relieved the pain in both legs. At autopsy, there was
an even greater posterior, but a lesser anterior, extension of the lesions on both sides.
There was no involvement of the posterior horns or posterior columns on either side.

317

318

Central Pain Syndrome

The differing incisions in these two cases intimate that the pain suppressor function
in the posterior columns was destroyed in the first case. Actually, posterior
cordotomies never gave rise to persistent pain (Chapter 2) and posterior columns are
intact in cases of CP following anterolateral cordotomies and ASAS. SCS, which
should engage this tonic suppressor mechanism, is ineffective even in the majority of
patients with retained DC function (see Chapter 6) and can even worsen CP (Triggs
and Beric 1994). The impairment of touch processing seen in CP patients can be
normalized by pain relieving procedures (e.g., propofol, TMS and CS), so that the
patient can feel touch normally (see also Nathan 1960). Actually, lemniscal activity in
BA3b and STT pain related BA3a are mutually inhibitory by intracortical connections
(Tommerdahl et al. 1996; see also Apkarian et al. 1994; Schnitzler and Ploner 2000).
In the human Vc, STT terminal clusters appear to be relatively separated from
lemniscal terminal areas, with some overlap (Lenz and Dougherty 1997). This
notwithstanding, the strongest opposition to a role of DC/ML fibers in CP comes
from several CP patients having intact epicritic conduction.
3. STT inhibits SRT. A slow multisynaptic spinoreticulothalamic pathway (SRT) is
strongly suggested by neurosurgical evidence (see King 1977; Gybels and Sweet 1989,
p. 192), but also by current clinical (medial medullary infarctions; Bassetti et al. 1997)
and neurophysiological data (Rousseau et al. 1999). According to several authors
(e.g., Noordenbos 1959; Hassler 1959; Cassinari and Pagni 1969; Nathan and Smith
1984; McGowan et al. 1997), damage to STT weakens this damping on SRT (slow
multisynaptic ascending system) at all levels of cord and brainstem and CP arises
(i.e., local disinhibition of the polysynaptic system which becomes hypersensitive):
The more the lesion spares the paleospinothalamic afferents, the greater the chances
of occurrence of central pain. (Lhermitte also postulated a damping of the SRT by
the ML.) At least one report (Mikula et al. 1959, 25 mesencephalotomies) concluded
that sparing the reticular system reduces the incidence of dysesthesias. On the other
hand, Amanos group (Table 7.1) abolished CP in many cases by selective destruction
of the reticular formation, with no new CP arising (unlike STT mesencephalic
interruption): this is the strongest evidence up to now of a role of the reticular
formation in CP.
4. Descending fiber damage. Some believe that the simultaneous involvement of
multiple different descending inhibitory fibers and ascending pathways may
be more important than denervation singly. An imbalance of descending modulating
pathways, affected at sites remote from the injury, is thought to be a mechanism
of release. It should also be recalled how there not only exist antinociceptive
systems, but also pro-nociceptive systems, so that different degrees of injury may tip
the balance toward one of them, favoring, or not, hyperexcitability and the onset
of pain (Millan 2002). Since the brainstem analgesia systems have bilateral effects,
damage thereof is unlikely to explain a unilateral pain syndrome; besides, it is
not clear what would predominate in CP, excitation or inhibition (Sandkuehler
1996; Porreca et al. 2002). Others find that initiation (but not maintenance) of
CP is independent of reduction or loss of descending or propriospinal inhibitory
fibers.

Piecing Together the Evidence

5. Thermosensory disinhibition. According to this hypothesis (Craig 1998), a cold

signaling enteroceptive A-delta STT path from spinal LI to thalamic nucleus VMpo
to (purported) thermosensory (cold-recipient) dorsal mid/posterior insula (which
then modulates brainstem thermoregulatory stations) normally inhibits a medial
heat-pinch-cold nociceptive (HPC) STT path (from multimodal cells receiving input
from C fibers) passing from LI through thalamic nucleus MDvc en route to ACC.
In CP patients, a lesion of the cold path disinhibits the medial path, with cold
allodynia and deep burning pain being selectively felt in the ACC, with activation of
homeostatic behaviors. Cold allodynia would be due to impairment of thermal
sensibility. This theory is totally refuted by an impressive number of observations:
(1) not all patients with CP complain of burning or thermally described pain; (2) a
minority only complains of cold allodynia, in the face of frequent (but not universal)
impairment of thermal sensibility, and actually the most extreme cold allodynia
occurred in a patient with normal cold detection thresholds in one study (Greenspan
et al. 2004); (3) disrupted thermal sensation in CPSP is not associated with a
corresponding relationship between altered cold perception and spontaneous pain
(Jensen et al. 2002); (4) CP may be felt superficially and in depth; (5) unexplained (by
the theory) nonthermal allodynia; (6) heat or increases of body temperature during
exercise or fever do not allay CP (as suggested by this theory), but in some cases (e.g.,
Romanelli and Heit 2004) may well worsen it (e.g., MS CP, plus heat allodynia in
some BCP/CCP cases): cool air actually temporarily reduced the intensity of CP in at
least one of our patients; (7) contrary fMR evidence of thermal coding in the insula
(Brooks et al. 2002); (8) no ACC activation during cold allodynia in CP patients
(Chapter 7, Section 1); (9) no unequivocal metabolic evidence for a role of ACC in
perceived unpleasantness of pain (Casey et al. 2001) and cold pain tout court; (10)
inefficacy of cingulotomies to relieve CP, but not other chronic pains; (11) clear-cut
anatomical evidence contrary to the existence and/or importance of the cold and
HPC paths (see complete list in Wall 1995); although a segregated warmth spinal
path exists in humans (Iannetti et al. 2003; Friehs et al. 1995). Lahuerta and
colleagues (1994) noted how surgical interruption of the STT does not abolish pain
sensation completely: only 1500 STT fibers reach the cortex, and other paths are
required; (12) non-exclusive role of VMpo as a thermal-specific thalamic relay
and ample doubts about its existence (Percheron 2004; Willis and Westlund 2004);
(13) failure of elicitation of painful or unpleasant sensations by electrical stimulation,
even with high currents, at sites in the ACC, where pain-sensitive neurons can
be recorded in human patients (Hutchison et al. 1999), as pain-related activity in
the ACC may represent descending modulation rather than perception of pain;
(14) a metaanalysis of all chronic pain imaging studies found that a decreased
incidence of activity in ACC and thalamus, coupled with decreased coding for
perceived pain in ACC, as well as an increased incidence of activity in the prefrontal
cortex, all contradict the thermosensory hypothesis (Apkarian et al. 2005).
6. Injury discharges. Rare patients have no sensory deficit, even with laser
SEPs. Frank injury is not necessary to induce the pain state: Indeed, the process
we call central sensitization may not require that a nerve be injured at all, but only
that a noxious stimulus be delivered (Devor et al. 1991). Enhanced patterns of

319

320

Central Pain Syndrome

electrical discharges can alter central processing of sensory input without

actual damage, provided an acute, severe pain stimulus is given. Even when injury
occurs, the first signals to reach the CNS and notify that it has taken place are
so-called injury discharges, short high-frequency signals lasting several minutes
at most transmitted along nociceptive fibers. It is speculated that a rapid depolarization ensues in dorsal horn inhibitory neurons, disabling them via an excitotoxic
mechanism, with attendant long-lasting disinhibition of primary afferent input to the
dorsal horn (the basis of preemptive analgesia: Bonicalzi et al. 1997).
7. Different side of injured brain. Right thalamic lesions are more likely to originate
CP (see Chapter 2). Increased frequency of chronic pain (including CP) after rightsided lesions may be due to right hemispheres specialization for negative higharousal emotions (including pain) and for monitoring of somatic state. In normal
subjects, pain threshold and pain tolerance are lower on the left hemisoma than the
right for electrical thermal and focal pressure stimulation; patients with right
hemisphere injury lesions tolerate pain longer than those with left hemisphere lesions
and show reduced galvanic skin responses to ipsilesionally administered painful
stimuli; there is greater alpha EEG suppression in the right compared to the left
hemisphere during exposure to thermal pain (see references in Nasreddine and Saver
1997). This may be due to cortically mediated attentional factors (Meador et al. 1998,
and references therein). Also, the human thalamus shows a strong right lateralization
of norepinephrine, which has a role in somatosensory information processing (Oke
et al. 1978; Canavero and Bonicalzi 1998b).
8. Intensity of STT damage. Since lesion location is not predictive of CP, as lesions
may be found at all levels from brainstem to cortex (opposite to speculations of other
authors, e.g., Sweet (1991), who believed the site of the lesion, rather than the
presence of the deafferentation, is the most critical factor determining pain onset),
it is speculated that the important factor in releasing CP is the degree of pain and
temperature loss and the level (concentration) of transmitter receptors in these
pathways (Bowsher et al. 1998). However, cordotomy severely or completely
interrupts STT, but CP follows in only a minority of cases, and so ASAS. Specularly,
CP patients with minimal damage to the STT exist.
9. Wrong temporospatial integration of sensory input (Kendall 1939: injury to fast,
but not slow, pain fibers eliminates refractoriness of end-stations to slow fiber input;
Walker 1955: greater interruption of STT fibers bound for Vc than for CL, with
diffuseness of output; Nathan and Smith 1984: massive cortical activation by
impulses traveling along the reticulothalamocortical paths of the diffuse projection
system). Subliminal impulses may become supraliminal, but since nociceptors do not
fire tonically (Ochoa 1993), only tactile stimuli should be involved.
10. Engagement of latent pathways. Injury appears to lead to the unmasking
of latent pathways: supposedly, central connections display a continuous adjustment
between activated and latent states, offsetting too much excitation or inhibition. This
may depend on inhibitory interneurons, keeping fringe inputs from being
effective. This unmasking would basically consist of a rapid reorganization due to
rebalancing of excitation/inhibition. According to Nathan and coworkers (1986),

Piecing Together the Evidence

the sudden onset of a lesion must alter the activity of the CNS . . . The usual
organization of facilitation and inhibition will be changed . . . The organization
of descending control is bound to be altered drastically. That there are possible
pathways constantly closed by inhibition was shown by Kirk and Denny-Brown
(1970) and Denny-Brown et al. (1973). In patients with complete section of the
spinothalamic complex and dense and persistent contralateral analgesia, immediate
onset of allochiria is seen in the previously analgesic area after cordotomy. Although
allochiria is not CP, it has supported speculations that CP may be subtended by an
already existing, but unavailable, subsidiary pathway (Nathan and Smith 1979;
Nagaro et al. 1993) localized in the posterior third of the cord (posterior horns or
Lissauers tracts), as reference of pain and CP may be observed following complete
destruction of the anterior two thirds of the cord (like in ASAS). This mirror pain (or
reference of pain or allochiria) occurs in 9
63.3% of cordotomized patients and in
some can be severe (Tasker and North 1997). Nagaro and colleagues (1993) found
that in 7/66 patients undergoing percutaneous cordotomy, allochiria appeared
immediately after induction of analgesia and disappeared when analgesia faded. It
was elicited only from analgesic areas and was experienced either at the same or a
more cephalad dermatomal level on the contralateral body (only in one patient in the
literature it was more caudad). Each patient with allochiria also developed new
contralateral, usually mirror, pain after cordotomy and, in one patient in whom
mirror pain appeared 6 hours after the cordotomy, subarachnoid phenol block
temporarily relieved both, suggesting a spinal mechanism. The fact that allochiria is
observed after destruction of the anterior two-thirds of the spinal cord (also in ASAS)
supports the idea that the pathways responsible for reference of pain, normally silent,
are found in the dorsal horn (lamina II) and connecting fibers (propriospinal)
including the tract of Lissauer. The disappearance of referred pain with contralateral
cervical cordotomy shows that the impulses causing reference of pain reach the dorsal
horn neurons which have a receptive field where the pain is felt and go up the
contralateral anterolateral column (Nagaro et al. 1993). Thus, allochiria and new
mirror pain may depend on loss of feedback inhibition from second-order neurons
and/or more central neurons of the nociceptive pathway of an already existing
subsidiary pathway, i.e., a network of short neurons which connects dorsal horn
neurons longitudinally and latitudinally. Destruction of tonic descending inhibitory
pathways is excluded, as reference of pain occurs only from the region rendered
analgesic by cordotomy, while disinhibition would occur more widely than this
region. Despite the interest of such rapid appearance of mirror pain, this is a
completely different phenomenon than acute-onset CP.

DISSECTING THE ROLE OF GABA

GABA is the most important inhibitory neurotransmitter in the human CNS.
Of three classes of receptors (A, B, C), GABA A receptors are foremost in the
regulation of brain excitability, timing-based signaling, setting the temporal window
for synaptic integration and synchronizing neural networks. However, it has

321

322

Central Pain Syndrome

become clear that there is no unique GABA A (and possibly also B) receptor in the CNS,
instead there is great heterogeneity, with a large number of different GABA A receptor
subunits with distinct regional distribution; also, individual GABA A receptor subtypes
are associated with distinct neuronal structures and subcellular distributions.
Thus, composition differs not only in different parts of the brain or in different cells,
but also in the same cell at different synapses, and their differential activation is
correlated with distinct pharmacological and behavioral phenotypes. The same
GABAergic axon can form distinct types of synapses onto different classes of target
neurons (Markram et al. 2004). While synapses at the soma control action potential
generation, synapses at distal dendrites control incoming input and propagation
of Ca2 currents. Each receptor subtype (more than 20) has its own target identity
depending on the subunits.
GABA receptors are pentameric heterooligomers; 19 distinct GABA A receptor
subunit genes are known, classified into 8 classes (a 1
6, b 1
3, g 1
3, d, e, y, p
and r 1
3). GABA A receptor assembly can be derived from a permutation and
combination of two, three, four or even five different subunits, with the majority
of subtypes in the brain composed of assemblies of alpha, beta and gamma
subunits. Distribution of the major subunits in various regions of the brain varies:
e.g., the cerebral cortex has intermediate levels of a 1
4 subunits and low levels
of a 5 subunit, whereas the thalamus contains high levels of a 4 subunits and
intermediate levels of d subunit (Olsen and Avoli 1997; Sleghart and Sperk 2002;
Kittler and Moss 2002; Rudolph and Antkowiak 2004; Olsen and Betz 2006).
The number of synaptic GABA A receptors can be dynamically modulated and the
modulation of 5HT and dopamine receptor function also hinges on modification
of GABA A receptor activity (Sleghart and Sperk 2002; Kittler and Moss 2002).
Finally, inhibitory interneurons are heterogeneous (about 50 types), some longranging, some also displaying bursting patterns; their role at both cortical and
thalamic levels may vary greatly among various types of processing streams (reviewed
in Markram et al. 2004).
The consequences of GABA A receptor activation on active membrane properties is context-specific, depending on the history of the membrane (the ratio of
activated to inactivated to closed voltage gated channels at the time of GABA A
receptor activation), the spatial location of GABA A receptors and the distribution
of voltage gated channels along the somatodendritic axis. GABA neurotransmission
can be excitatory in basal conditions, also in adult tissue, so that hypofunctionality
of a GABAergic pathway can in reality decrease the global excitability of
the network. Unlike cationic glutamatergic synapses, GABA synapses have
a unique feature resulting from their chloride permeability that enables them
to shift from an inhibitory mode of operation to one that mainly excites (Cossart
et al. 2005).
What happens following injury (for simplicity, we assume no fundamental
difference among trauma, ischemia and compression)? Considerable and rapid
plastic changes in the amount and distribution of many CNS receptors take place:
neurons can alter their chemical code and increase, decrease or change the expression
of neurotransmitters; intense electrical activity may in itself affect the concentrations
of some peptides (see Bullitt 1991). However, data strongly suggest that the key

Piecing Together the Evidence

to CP lies in GABA receptors, specifically GABA A receptors. Given its clear-cut

profile of action at subhypnotic dosage, propofol provides an important window,
much more than barbiturates or benzodiazepines (see Chapter 5). GABA A
modulation affords the strongest levels of CP relief (particularly propofol)
while
ketamine, less potent, would point to NMDA receptors
and appears to be
more effective than GABA B modulation (although this assumption is based on
incomplete evidence).
Following injury, not only the number of GABA A receptors can change, but
even, and perhaps more importantly, the subunit composition (subunit switch)
as
also seen in epilepsy
perhaps as a result of aberrant compensatory plasticity
(i.e., a new receptor isoform that is less functional might become relatively more
abundant). Intracellular changes other than subunit switch can also change GABA
A receptor function: receptorial or release mechanisms are primary candidates
to explain those alterations, e.g., phosphorylation processes or anchoring of the
receptors by gephyrin, GABA receptor-associated protein and others. Uncoupling
may be rapidly produced and reversed, without alteration of gene expression.
The result may be altered, persistent, regulation of inhibitory function, i.e., hypofunctionality. The process of subunit switch is clearly dynamic (Cossart et al. 2005).
Both SI and sensory thalamus show a tonic inhibitory tone, modulated by sensory
input: this has important consequences on stimulus localization and receptive
fields (RFs), but also for compensatory adjustments following injury. In the
cortex, GABA has a particularly high density in layer 4. Prolonged or overactive
GABAergic synaptic transmission (or chronic high doses of GABA agonists such as
benzodiazepines and baclofen) can lead to decreased (downregulated) GABAergic
function (A and B). This plasticity may occur over a few seconds (e.g., during
modestly enhanced glutamate activity)
up to years in other contexts. GABA levels
in the human SI are reduced within minutes of deafferentation (Levy et al. 2002) and
propofol data point to a disrupted GABA A inhibition. Subtle reductions in GABA
inhibition result in large changes in excitatory conduction and spread of activity to
distant cortical sites (even in the face of paradoxical increases in evoked polysynaptic
inhibition due to enhanced excitatory drive onto GABA interneurons) and local
changes in GABA activity may lead to temporary associations between adjacent
cell groups, enabling reorganization (Jacobs and Donoghue 1991). Also, a deficit in
dendritic (versus somatic) inhibition (from dendritic projecting interneurons) can
reduce the excitability threshold (Cossart et al. 2005). Inhibition is carefully
modulated at several levels, including specific transporters and tonic spill-over
currents (Semyanov et al. 2004) in a dynamic balance.
The b subunit (b 2-3) is key to the direct actions of propofol (reviewed in Rudolph
and Antkowiak 2004). Low versus high doses differentially activate different subunits, and propofol may directly activate GABA A receptors in the absence of
GABA (reviewed in Olsen and Avoli 1997). Barbiturate and benzodiazepines
action, in contrast, hinges on a subunits (Olsen and Avoli 1997; Rudolph and
Antkowiak 2004). These data may guide the search of
possibly
altered genes
in CP patients.
Exceptional cases of patients worsened by GABA agonists (Chapter 5) are not
in contradiction to the above discussion. Inhibition is not always reduced: for

323

324

Central Pain Syndrome

instance, in a few brain regions of some epileptic patients it can be increased as

a compensatory mechanism to hyperexcitation (Brooks-Kayal 1998; Nusser et al.
1998). Moreover, GABAergic activity may be excitatory depending on local circuitry
through a disinhibition (Koehling 2002); also, when GABA B receptors are located
at inhibitory terminals (autoreceptors), their activation will decrease GABA release
and may result in excitatory influence (Olsen and Avoli 1997). Hyperactivity at
GABA synapses leads to an increased number of postsynaptic GABA A receptors
and an alteration in their subunit composition (Olsen and Avoli 1997). Paradoxically,
small decreases in glutamatergic excitation within cortical circuits which may
decrease excitation of interneurons might also lead to a relative disinhibition of
pyramidal neurons and hyperexcitability (Salin and Prince 1996). More simply,
in propofol non-responders (generally patients with ischemic/mechanical disruptive
rather than compressive lesions), the hypothesized glutamatergic hypertonus may
be too intense to be swamped (see in Canavero et al. 1996).
In sum, following injury, alterations occur rapidly, do not necessarily require
neuronal damage, may become long-lasting; an inhibitory-to-excitatory shift of
GABA actions can even become permanent (Cossart et al. 2005).

WHICH ROLE FOR NEUROPLASTICITY?

Neuroplastic changes are at the basis of much current thinking on the pathogenesis
of chronic pains of different kinds and these will be shortly discussed.
That injury may cause hyperactivity has been known since Hall (1841) and
Claude-Bernard (1880). A supersensitive state of denervated neural structures was
postulated by Cannon and Rosenblueth (1949), who, even though they did not
mention pain, recognized the fact that reorganization of the CNS follows
denervation, discussed plasticity of the neurons, the interchangeability of the
nervous pathways and influences, the functional opening of new vicarious
pathways determined by training and the onset of initiative foci.
Supposed animal models of chronic neuropathic pain evidenced a smorgasbord of
plastic alterations, all apparently important in their genesis, that are rather difficult,
even for proponents, to integrate in a coherent picture, including possible preemptive manipulation (e.g., Juliano et al. 1991).
One commonly discussed entity is so-called central sensitization, which follows
prolonged or repeated noxious stimulation of STT neurons at the time of the
pain-inducing lesion, possibly due to loss of effectiveness of inhibitory mechanisms
within the spinothalamocortical pathway. This consists of a spectrum of derangements which include increased spontaneous discharge, evoked pains, but also
denervation supersensitivity (an enhanced response of neural cells to the transmitter
lost, and then reexpressed) and RF expansion, due to loss of sensory input. It is
considered to be a form of long-term potentiation. So-called wind-up, a progressive
increase in neuronal excitability akin to sensitization, observed in a minority of
spinal cells only and some, but not all, chronic pains, follows repeated stimulation
of nociceptive C fibers (Baranauskas and Nistri 1998). Although the NMDA
receptor is considered pivotal to such changes, long-term sensitization actually

Piecing Together the Evidence

requires co-activation of several receptor systems and possibly complicated unsuspected mechanisms, depending on species and cell type, among others (Baranauskas
and Nistri 1998); NMDA receptors are involved in many forms of synaptic plasticity, so that additional mechanisms are necessary to impart specificity to paininduced sensitization (making prevention of sensitization through NMDA block
impractical).
Another plastic change which has been amply discussed as a possible contributor
to chronic neuropathic pain is sprouting, a hierarchical and lesion-specific, nonrandom phenomenon, which follows injury. It includes collateral sprouting from
uninjured neurons with variable restoration of anatomy (rapid: days, complete in
days or months; extent: 40 micra), ingrowth from healthy, but functionally distinct
neurons (1 month after injury), pruning (with growth of new axons from injured cell)
with a more normal anatomic restoration (4 months after injury, lasts 2, extent up
to 1 mm). Sprouting can lead to rewiring: e.g., intracortical sprouting can lead
to generation of powerful monosynaptic excitatory feedback and intraspinally Ab
fibers may retarget STT neurons, or non-pain-relaying neurons, in LII and even
switch neurochemical profile and function as C fibers. At the end of a literature
review, Tasker and Dostrovsky (1989) concluded that, if sprouting occurs, it is of
very limited extent and probably limited to a subpopulation of primary afferents
and/or axons of CNS neurons, playing no role in receptive field expansion.
We have already discussed decreased inhibition from lesions of descending or
segmental pathways, with changes in facilitation and inhibition (see above).
A fundamental role is attributed to somatotopographical rearrangements (representational remodeling), with expansion of RFs at all CNS levels, supposedly
due to changes in synaptic efficacy, disinhibition with unmasking or strengthening of latent, but ineffective, excitatory and convergent synaptic inputs, changes in
intracellular processes leading to altered neuronal excitability, sprouting with
creation of new synaptic connections. Acute lesions as well as manipulation of
sensory inputs, can lead to rapid reorganization of the cerebral cortex, occurring
within minutes to hours (references in Levy et al. 2002), through a rapid reduction
of tonic GABA inhibition. Somatotopic rearrangements (up to 2
3 cm in cortex)
have been reported in human pain states, namely phantom pain, and widely believed
to correlate directly with painful sensations, particularly at cortical levels (Flor
2003). The more extensive SI reorganization after injury may depend in part on the
activation of the widespread network of horizontally connecting axons within cortical
areas
a feature missing in subcortical areas (e.g., the thalamus); the immediate
expansion or new expression of RF in SI (disinhibition of silent inputs from
body areas adjacent to denervated areas) may be subserved by the wide arborization of TC afferents. Cortical layers contribute differently to plasticity: cells in
supragranular and infragranular layers respond rapidly to changes in sensory
experience and then contribute to modifications in Layer 4 (Diamond et al. 1994).
Yet, short-term dynamics of horizontal pathways in the middle of uniformly
deprived SI change only modestly and vertical intracortical pathways are unaffected
following loss of input. Thus, uniform loss of sensory activity has a limited effect
on short-term synaptic dynamics and competition between deprived and spared
sensory inputs is necessary to produce large-scale changes in synaptic dynamics

325

326

Central Pain Syndrome

after sensory deprivation (Finnerty and Connors 2000). Human evidence disproving
the role of somatotopic rearrangement has been published (e.g., Moore and
Schady 2000; Vega-Bermudez and Johnson 2002) and referred sensations/
mislocalization do not appear to be a direct perceptual correlate of cortical
reorganization (Knecht et al. 1996). Knecht and colleagues (1998) report that
phantom sensations can be evoked even in normal persons without deafferentation
and pain itself in chronic pain patients can lead to representational reorganization
(references in Knecht et al. 1996).
Generally discussed with injury-related neuroplasticity is neuronal degeneration,
including inhibitory cells, and shrinkage, along with substantial dendritic atrophy,
loss of dendritic spines and truncated dendrites and/or loss of the proximal axons
and perikaryo-nuclear alterations. However, transneuronal degeneration with
neuronal loss may be incompatible with concurrently extant central sensitization,
and acute loss of GABA cells, for one, is excluded by the timeline of GABA decrease
(although GABA cells appear to be particularly sensitive to disruptions of blood
flow to the brain: over time these effects might kill them or reduce their ability
to make and release GABA).
Truth is, neuroplasticity is something intrinsic to the nervous system, independent
of injury. For instance, cortical maps express experience-dependent plasticity and
SI normally reorganizes during various tasks; after injury, it serves a purpose of
recovery.
All the discussed neuroplastic changes simply cannot sustain irreversible changes
during CP, because CP can be promptly reversed (Chapter 7). This is not unique
to CP. Cases of years-long neuropathic pains, including trigeminal neuralgia and
carpal tunnel syndrome pain, resolve immediately after pain relieving surgery
(Schott 2001a). Just as chronic pain is so often entrenched, changes in the nervous
system following injury and disease (central sensitization) might be envisioned as
irreversible. Abolition of chronic pain resulting from gross structural nerve damage
sustained over many years is difficult to explain by current views on the major and
extensive peripheral and central somatosensory changes thought to occur after nerve
lesions. Interestingly, reversible epidural blocks of the nerve roots result in the
acute appearance of new RFs that are lost and replaced by the original RF after
the peripheral nerve recovers (Metzler and Marks 1979), pointing to great flexibility.
Thus, either that they can be rapidly reversed or not, such changes would be
inconsequential. Instead, they might play a role in the initial stages of CP: a GABA A
receptor subunit switch was previously suggested. Further, loss of sensory input
cannot explain immediate and delayed-onset pains, which are clinically identical:
in the former, processes involving slowly developing, continually progressive
neuronal changes cannot be essential for the generation of pain; likewise, loss of
sensory input produces an immediate and simultaneous change in neuronal activity
at multiple CNS levels
for instance, human thalamic neurons develop novel RFs
within minutes (5
15 minutes) of lidocaine block (Kiss et al. 1995) and SI rapidly
reorganizes during acute cluster headache attacks (Soros et al. 2002), with dendritic
filopodia appearing within minutes (Maletic-Savatic et al. 1999). Thus, it is difficult
to understand the progressive entrenchment (Schott 2001) some believe to exist.
Further, denervation supersensitivity is present in both pain and non-pain cases.

Piecing Together the Evidence

In the nontraumatic cervical anterior spinal artery syndrome, a relatively rare

anterior myelopathy with severe, practically complete interruption of the STT at the
spinal level, STT fibers cannot be involved in any kind of transmission from the
periphery, and thus maintain sensitization (Beric 1993) (however, uninterrupted SRT
projections might play this role).
According to Taskers group (Kiss et al. 1994), the role of somatotopic reorganization in the genesis of CP
but also PNP
is entirely speculative. Unlike animal
models, there appears to be different patterns and degrees of somatotopic
reorganization in the human, all (or none) of which may be associated with a
pain syndrome. They conclude: Although in some cases changes in somatotopic
representation were observed, these changes were not consistent in all the groups
and therefore unlikely to be the common cause of pain in these patients. Ojemann
and Silbergeld (1995) found that adult human sensory cortex retains its somatotopy
even after two decades without conscious perception of that body part, after major
peripheral denervation
unlike MI. Woolsey and colleagues (1979) also found
maintenance of cortical sensory maps. Experience with extradural cortical
stimulation in CP (see Chapter 6) confirms that sensory maps (the homunculus)
are stable. Unlike many, but not all, primate models of SI plasticity, humans display
a relative preservation of the cortical sensory homunculus. Thus, in humans,
deprived, but reactivated neurons do not take on new and appropriate functions, but
carry out their original roles long after they have had time to adopt new ones (Davis
et al. 1998). In a study of 12 thoracic SCI patients, 9 reported phantom sensations
and 2 referred phantom sensations. In these 2, fMRI showed a relation between SI
activation and the percept of referred phantom sensations. The authors concluded
that, instead of somatotopical cortical reorganization, cortical plasticity may be the
expression of co-activation of nonadjacent representations even distant between
them, supported by somatotopic subcortical remapping projected to the cortex
(Moore et al. 2000). Turner and colleagues (2003) examined with fMRI a group of
SCI patients versus healthy controls. Unlike amputation, no evidence of expansion of
the hand representation into nearby cortical areas was found, with hand sensory
representation undergoing a much smaller posterior shift of hand motor
representation. Reorganizations in the order of those seen in phantom sensation
simply lack in CP. However, those few cases of CP regressed after stopping MCS (see
Chapter 6) have been explained with some kind of reverse neuroplasticity in SI.

BILATERALITY OF CENTRAL PAIN

Although individual brain regions and networks of brain regions exhibit some
degree of functional specialization, acute pain is processed by a highly distributed,
redundant and resilient brain system, with
unlike other sensory modalities

detailed information about the intensity of a painful stimulus being conserved
at multiple levels in both hemispheres
an arrangement that would justify the
difficulty of eliciting painful sensations by cortical stimulation, as simultaneous
activation of several regions may be necessary (Willis and Westlund 2004, and
references therein). In other words, there is no unique pain center. Acute pain

327

328

Central Pain Syndrome

is essential for survival: individuals born without the ability to perceive pain
frequently die from injuries and infections they have never felt. The distributed
processing of pain within the human brain ensures that this critical ability to detect
tissue injury can be spared in the face of extensive CNS damage. On such grounds,
Melzack-Caseys hypothesis of sensory discriminative and affective motivational
components of pain being processed in parallel by distinct neural systems fatally
collapses, all the more so since it has never been seriously tested (Fields 1999).
However, this is not equivalent to saying that chronic pain cannot be effectively
abolished by selective lesions: while acute pain is necessary for survival, chronic pain
is not, and abolition of even focal generators can relieve it. The impression that
chronic pain cannot be abolished by focal lesions is due to poor analysis of the
relevant literature and misconceptions about the exact generator of a particular
chronic pain syndrome, as in the case of CP. On the other hand, the same neural
substrates that support the bilateral distribution of nociceptive information
processing during acute pain could subserve bilateral spread of chronic pain.
Somatosensation may be served by ipsilateral brain structures (reviewed in Coghill
et al. 1999), as shown by hemispherectomy cases (Dandy 1933; Muller et al. 1991).
Noordenbos and Wall (1976) described a patient with spinal cord transection with
saving of only a part of one anterolateral quadrant, who could perceive tactile and
painful stimuli on both body halves. Patients with corpus callosum transection
can report tactile and painful (Stein et al. 1989) stimuli from either body half.
Contrary to traditional views of spinothalamic transmission of pain, a significant
proportion of functional imaging studies (8/20) employing unilateral painful stimuli
have detected activation of both the contralateral and ipsilateral thalamus (see
references in Coghill et al. 1999), which cannot be understood as a generalized
arousal in reaction to pain.
Following PNS lesions, there are well-documented events affecting the contralateral nonlesioned structures, qualitatively similar to those occurring on the
ipsilateral side (although usually smaller in magnitude and with a brief time course):
mirror pains contralateral to nerve injury in humans have repeatedly been noted,
and so bilateral hyperalgesia after unilateral nerve injury (Mohammadian et al. 1997,
Oaklander et al. 1998); a significant percentage of patients with so-called complex
regional pain syndrome experience bilateral spread of pain despite an initial,
unilateral injury (Veldman and Goris 1996); SCS, like MCS (see Chapter 6) can
induce bilateral effects following unilateral stimulation (Lazorthes et al. 1978;
Garcia-Larrea et al. 1989).
This is also the case for CP. Riddoch and Critchley (1937) reported exceptional
cases of bilateral pain due to unilateral thalamic lesion. We (Canavero 1996)
described a woman with a subparietal cavernoma and contralateral CP who, for
about 10 days, complained of the same kind of pain (burning paroxysms to arm
and, when severe, the whole hemisoma) on the contralateral arm. Both pains
simultaneously responded to propofol. No sensory deficits were ever observed in
involved areas. Kim (1998) described six patients with unilateral stroke who
initially developed painful sensory symptoms on the side contralateral to the lesion.
The patients CPSP progressively worsened for a certain period of time when
sensory symptoms also occurred on the side ipsilateral to the lesion. The delayed

Piecing Together the Evidence

onset ipsilateral sensory symptom was mild, unaccompanied by objective sensory

deficits and developed in the body parts mirroring the site of the most severe
CPSP. Once developed, they persisted during follow-up (new-onset PNP and
strokes were excluded by appropriate exams in some patients). We have already seen
examples of bilateral CP elicited by unilateral stimulation in previous sections
(Gorecki et al. 1989; Chapter 7, Section 2). Kim (1999) also reported on five patients
with hemisensory symptoms due to unilateral strokes occurring in the left
putamen, left thalamus, right putamen, right lateral medulla and left thalamicinternal capsular area. Sensory symptoms had gradually improved or remained
stable after onset. When another stroke occurred on the contralateral thalamicoccipital, frontoparietal, lateral medulla, temporoparietal and pontine areas,
respectively, previous sensory symptoms significantly worsened and became painful
on the previously affected side. Also, two patients with sudden remission of CP
following a new stroke in the unaffected hemisphere are on record (see Chapter 7,
Section 4). Taskers group described cases of CP patients with a silent thalamus,
who most likely engaged the healthy contralateral hemisphere (see Chapter 7,
Section 2). Greenspan and colleagues (2004) had 2/13 cases of unilateral CPSP with
bilateral cold hypesthesia and other studies described a small number of patients
with bilateral cold hypesthesia (Beric et al. 1988; Boivie et al. 1989; Vestergaard et al.
1995). Thus, one fact seems inescapable: the mechanism that leads to CP engages both
hemispheres, so that a corticothalamic pain loop can be activated on either side;
importantly, bilateral CP does not depend on structures with bilateral receptive fields
(e.g., SII or ACC) or contralateral strokes would not abolish the pain. CP may
likely be shifted contralaterally through the corpus callosum (transfer time: 15 ms;
Frot and Mauguie`re 2003) or through the reticular formation, including spinal and
brainstem commissural interneurons (Koltzenburg et al. 1999), with its bilateral
projections, following loss of GABA tone in SI and reticular formation priming
(see previous discussion). Olausson and colleagues (2001) found that cortical areas
typically involved in pain processing can be activated by ipsilateral pathways directly
from the periphery, but, unlike tactile information, pain activation in the hemisphere
contralateral to the stimulation is dependent on transcallosal information processing.
In amputees, acute hand deafferentation can elicit a focal increase in excitability
in the hand motor MI representation contralateral to the deafferented cortex that
is influenced by transcallosal interactions; GABA A agonism blocks this increased
excitability (Werhahn et al. 2002). Meyer and colleagues (1995, and references
therein) found that homotopic regions of SI are linked, so that plasticity induced in
one hemisphere (in the form of RF expansion brought about by a small peripheral
denervation) is immediately mirrored in the other hemisphere: neurons which
displayed the plasticity showed no responsiveness to stimulation of the ipsilateral
body surface, suggesting a specific role of maintaining integration between
corresponding cortical fields. Excitation may be followed by inhibition when the
stimulated area is larger or stimulation strength higher (see also Calford and
Tweedale 1990). Bilaterality of hand representation in parietal somatosensory areas
is under callosal control, since it is lost after callosal section, mostly at BA2 (but
much less at BA1 and almost none at BA3b) and BA5/7 levels (Iwamura et al. 1994).

329

330

Central Pain Syndrome

Since facilitatory interhemispheric influences are possible in patients with agenesis

of the corpus callosum, both mechanisms (corpus callosum transfer and reticular
formation-processed switch) may play a role.

LESSONS FROM CORD CENTRAL PAIN

For almost 50 years, a dichotomy of response between episodic and constant pains
in the setting of spinal injury has been discussed. Botterell and colleagues (1954)
stated: Burning pain has proved a problem difficult of solution . . . By contrast,
jabbing, shooting, crampy, gripping, colicky and vice-like pains, have been regularly relieved by bilateral tractotomy. Porter and colleagues (1966) wrote: The
effectiveness of cordotomy in relieving the symptoms of sharp, lancinating pains
in the lower extremities in patients with cauda equina lesions is summarized . . . The
operation had no effect, however, on the frequently encountered burning pain
in the lower extremities. White and Sweet (1969), seemingly inferring that
intermittent pain is radicular in origin and steady pain of central origin,
concluded: Cordotomy is very useful in paraplegia for relief of pain of radicular
origin . . . Provided the injury involves the cauda equina and does not extend rostrally
beyond the conus medullaris to involve the cord, we believe that relief can be
obtained in a high proportion of cases by anterolateral cordotomy. Jefferson (1983)
noted that cordectomy was differentially effective for discrete pain radiating into
the thighs, knees or legs, especially if shooting and episodic, and especially if caused
by lower cord lesions. Pain associated with high lesions, particularly if diffuse,
steady and in a bathing trunks distribution, was relieved poorly. He stated: One
of the very interesting, and perhaps characteristic features of the pain which is likely
to respond . . . is that it is episodic. Tasker and associates (1992) found a statistical
correlation between disappearance of intermittent and evoked pain and demolitive
procedures, compared to these latters ineffectiveness for spontaneous pain. They
wrote: destructive surgery is selectively successful in relieving the spontaneous
intermittent, often shooting radicular pain that tends to project down the legs . . .
present in 30% of . . . patients with cord central pain . . . particularly associated
with thoracolumbar lesions . . . evoked pain, present in 47% of . . . patients, responds
similarly to destructive surgery. Intermittent and evoked, but not steady, pains
should be dependent upon transmission in somatosensory (probably spinothalamic)
pathways, intermittent shooting pain perhaps being the result of ectopic impulses
instituted at, or proximal to, injury sites (e.g., through ephapses or peripheral ectopic
pacemakers) and then transmitted centrally in these pathways to be perceived as
pain. Pagni and Canavero (1995) also noted that the paroxysmal components,
often associated with spasms, usually due to lesions at T9
T12 vertebral level,
are satisfactorily relieved by cordomyelotomy. On such basis, it has been concluded
that evoked pains depend on a local cord generator, whereas diffuse steady pains
on more rostrad stations. According to Tasker (2001), intermittent shooting (89%)
and allodynia-hyperpathia (84%) respond to cordotomy-cordectomy-DREZ; steady,
causalgic, dysesthetic, aching pain only in 26% of the cases.

Piecing Together the Evidence

Another dichotomy has been noted between end-zone or girdle (at-level) pain
and diffuse (below-level) pains. On the basis of results of conventional DREZ surgery
and cordectomies (Chapter 7), it was concluded that steady burning pain referred
to the lower abdomen, and burning or dysesthetic pain diffused to the legs or
localized to the retroperitoneal region, buttocks or feet usually are not relieved.
Best results were reported for patients complaining of shooting, paroxysmal pain
(and spasms), even though referred to apparently totally anesthetic and paralyzed
limbs, and girdle pains. Pain worsened by bowel or bladder distension was also
likely to be improved by surgery. Results of cordectomies have been less rewarding
with lesions and sections at levels higher than T10.
Pain relief in paraplegics after cordectomy appears to be directly related to the
extent of the removal, with better results occurring when long rostral segments
of the cord are resected, that is, 2
3 cm (three spinal segments) are resected
above the site of injury (e.g., Druckman and Lende 1965; Table 7.1). Loeser and colleagues (1968) pointed to the cord segments rostral to injury playing important
role in the genesis of pain. Jefferson (1983) noted that, although abnormal tissue
was left above the level of his resections, without apparently influencing pain relief,
sometimes extension of cordectomy to apparently normal tissue was necessary.
Bilateral DREZ lesions that involve two to three spinal cord segments above the
spinal injury, and extend into normal cord, achieve a better pain relief (coagulation
includes laminas I
IV, but may involve up to lamina VI and adjacent white
columns), as damage extends for several segments well above injury site (Nashold
1991), whereas extension of DREZ lesions caudad into the sacral segments of the cord
does not improve the results (only 1 patient with diffuse sacral pain improved in the
series of Friedman and Bullitt [1988]; Table 7.1).
Edgar and colleagues (1993) and recently Falci and colleagues (2002) (Table 7.1)
found that DREZ surgery can indeed relieve diffuse pains, if lesions are extended
sufficiently. In the latter paper, in 62% of patients with below-level pain, spontaneous
DREZ hyperactivity was found 3
5 levels cephalad to injury level (7 in the series
of Edgar et al.). Their findings contradicted traditional dermatomal mapping
and thus they hypothesized that below-level pain was mediated significantly by
interneuronal pathways, while at-level pain was assumed to be mediated through
more traditional pain pathways (e.g., STT) corresponding to the DREZ at injury
level. Spinal block studies also (see Chapter 6) found that block above lesion level was
necessary for analgesia; failure in two patients (both with below-level pain) despite
anesthesia two levels cephalad to injury supports even more rostral mechanisms.
Davis and Martin (1947) wrote: If the distal end of the proximal segment of the
injured spinal cord was anesthetized by spinal anesthesia, the pain disappeared
(p. 493), This suggests that the origin of the pain was the end of the proximal
segment of the injured spinal cord . . . operations upon the sympathetic nervous
system [being] ineffective. Studies (Finnerup et al. 2003a,b) indicate that changes
in somatosensory function in dermatomes rostral to the injury level may be
important in the sustenance of CCP, with a significant correlation between intensity
of brush-evoked dysesthesia at lesion level and spontaneous below-level pain.
The same group (Finnerup et al. 2003c) also examined 23 SCI patients above T10
(14 with CP and 9 without CP) in an MRI study. At the level of maximal cord injury,

331

332

Central Pain Syndrome

21 patients had lesions involving the entire cord on axial images, except for a small
border of lower signal intensity, whereas 2 patients had central lesions. Rostral to the
main injury, the first image with an incomplete lesion showed significantly more
involvement of gray matter in pain than in pain-free patients. According to Defrin
and colleagues (2001), both a critical level of injury and a state of hyperresponsivity
is necessary for CCP to arise. Thus, apparently, above-level hyperactivity may
sustain CCP.
This diffuse, likely bilateral, spinal generator involving multisynaptic propriospinal systems in and around the lesioned gray matter may feed the thalamocorticothalamic loop; a similar generator would be present in the brainstem reticular
formation. However, this hyperexcitability is useless without STT-induced changes
at corticothalamic levels, which is the first step needed for CP to arise.
In thoracolumbar lesions, further excitatory input may derive from peripheral
(root/nerve) mechanisms. This hyperactive core may have variable extent
depending on subject. It is important to remember, though, how previously
unreported burning sensations developed after cordectomy (Botterell et al. 1954) and
even Falci and colleagues, who believed that the higher temperature they used had
markedly decreased the development of new squeezing, pressure pains, possibly
because of a more complete destruction in deeper laminas, triggered new CP
sensations; moreover, they could not relieve all their patients of below-level CP,
implying even more rostral hyperactivity. Beric (1993) pointed out that the ASAS
syndrome is characterized by severe, practically complete interruption of the STT at
the spinal level: here, the hypothesis of dorsal horn nociceptive cell hyperactivity at
the level of the lesion becomes inconceivable and useless in explaining the painful
symptoms of this syndrome. However, propriospinal hyperactivity can still be
present and hyperexcitability may have spread more rostrally.
Previously reviewed studies suggest that CP is much more frequent in incomplete
cord injuries. Actually, a majority of seemingly clinically complete transection injuries
are subclinically incomplete and retain significant communication between segments
above and below the cord injury zone even many years after the original trauma, as
shown both anatomically and electrophysiologically (Dimitrijevic 1987; Beric 1999),
so-called dyscomplete lesions, i.e., no volitional sensorimotor functions below
lesion level, but some residual descending function or control demonstrated by
electrophysiology. Also, some sensory cortical evoked responses may still be
detected in SCI patients with no clinically appreciable sensory function below the
lesion site; prolonged, repeated or continuous application of different stimuli may
be transmitted from below lesion level to the brain and produce the awareness
that something is happening in seemingly anesthetic areas (this is the case of
peripheral or central pathways still transmitting across the traumatic lesion on
fast or slow conducting fibers which are, however, functionally useless (Donovan
et al. 1982)). Finnerup and associates (2004) compared 24 SCI patients (11 with
CP and 13 without) with a clinically complete SCI (ASIA grade A), and found
that painful or repetitive pinprick stimuli elicited vague localized sensations in
50% of cases. SEPs and MRI found no difference between groups. Thus, sensory
communication was retained across injury level (sensory dyscomplete SCI).

Piecing Together the Evidence

Kakulas and colleagues (1990, and references therein) observed that out of 197 SCI
cases, only 22 reported pain and 5 burning sensations. Of these, 18 had clinically
incomplete and 4 a complete cord transection syndrome; in 10 cases the lesion was
cervical, in 6 thoracic and in 6 lumbar. They concluded that: there is a larger
proportion of patients with pain and abnormal sensations with anatomically
incomplete injuries. They also noted that an extensive regeneration of nerve roots
at the level of injury is more frequently observed in patients suffering from pain
and that most of seemingly clinically complete cord transection syndromes (63 out
of 88) show, on pathological examination, continuity of nervous tracts across the
lesion, with a variable residuum of descending and ascending central nervous fibers
running in the wall of the lesion. They also noted that spinal cord lesions are spread
over many segments below and above the level of the bony lesion and lesions may
extend well above the injury site (Durward et al. 1982). At these levels, loss of
myelinated fibers and neurons of the gray substance and gliosis intermingle. Damage
of the roots may swing from minor damage to complete or nearly complete
loss of nerve fibers. In the chronic stages, nerve root regeneration is a typical
feature with formation of neuromas, and is more frequently seen in cases of pain.
In traumatic spinal cord damage, the end result is a scar, with collagenous connective
tissue and, in the less damaged parts of the cord, demyelination of fibers, intense
astrocytic fibrous gliosis (Hughes 1976), involving posterior and anterior horns,
plus schwannosis. Surviving axons in injured spinal cord (MS, cervical spondylosis:
Hughes 1976; extramedullary tumors: McAlhany and Netsky 1955) have neurophysiological features typical of demyelinated axons (Rasminsky 1980). Then, sensory
loss would be due not so much to loss of axons (both during trauma and MS), but
to loss of their ability to transmit properly encoded information, conducting slower
and ineffectively. Repetitive discharge results in a change of the axons membrane
excitability caused by the propagation of an impulse per se (activity-dependent afteroscillations). Failure of sensation may occur with such activity-dependent changes
in axonal excitability, justifying the periodicity of intermittent conduction which
occurs in spontaneous discharge in axons with focal demyelination and in
spontaneous trains of discharge from axons ending in neuromas (Thalhammer and
Raymond 1991). Demyelination may involve cord tracts under the compressing
lesion or on the opposite side of the cord.
Pathological afferent discharges may spontaneously originate in the surviving
central stumps of divided central nerve fibers and damaged demyelinated fibers
of both anterior and posterior cord quadrants, with impulses arising ectopically
(Smith and McDonald 1982), both in incomplete and dyscomplete spinal cord
traumatic transections. Demyelinated axons may be responsible for pain paroxysms
(Pagni and Canavero 1993); minimal mechanical deformation of the cord at the
lesion site both increases the level of previous spontaneous activity, inducing
spontaneous activity in silent fibers.
Abnormal excitability is likely a general attribute of any demyelinated central
nerve fiber, including dorsal columns. These may be at the basis of Lhermittes sign

which would then be a typical mechanical irritative phenomenon of the posterior
columns at cervical level
and other electric shock-like pain paroxysms with
lesions inside the cord or compressing it from outside. Interestingly, such painful

333

334

Central Pain Syndrome

fits may depend on lesions of the dorsal columns, where visceral pain fibers are
found (Pagni and Canavero 1993; Willis and Westlund 2004). Here, cordotomies
would be ineffective. That scarring-induced irritation of sensory pathways might
play a role in CP was surmised by earlier authors: constant bombardment by
subliminal impulses from the periphery that under normal conditions are not felt
produces painful sensations (Garcin 1937). Traumatic lesions of the spinal cord
and cerebral cortex give rise to acute terrible aching pain which fades away in
a few hours, days or weeks (Garcin 1937). However, lesions that interrupt central
pathways will result in wallerian degeneration of the axons, and thus there is no
way for interrupted central axons to became a source of ectopic nerve impulses,
as can happen with peripheral axons, for example, in neuromas (Willis 1991).
In traumatic lesions at the thoracolumbar passage (T10
L1), which generally
involve both the cord lumbar enlargement and caudal roots, neuromas form in
the chronic stage (Kakulas 1990; Pagni and Canavero 1995). Nerve root lesion
or compression may add to denervation and ectopic impulse generation inducing
and maintaining a hyperexcitability of central multimodal neurons. From neuromas,
which are powerful ectopic pacemakers (though the amount of spontaneous
electrical activity of neuromas may have been overestimated; Burchiel and Russel
1987), and dorsal root ganglion cells of damaged roots, abnormal spontaneous,
paroxysmal discharges may sustain the activity of partially deafferented multimodal
neurons in the dorsal horns of not completely destroyed cord segments and
in the nearby less damaged rostral cord segments, at- and above-level and not
necessarily from a change in the excitability of the spinal neuron itself (Pagni and
Canavero 1995).
A role of Lissauers tracts, which lie outside the area included in anterolateral
cordotomies, must also be envisioned in the rostral spread of hyperactivity.
Denny-Brown and colleagues (1973) found that the medial division of Lissauers
tract seems to exert a facilitatory effect, and the lateral division a suppressor effect
on transmission of afferent impulses at the first synapse. Lesion of the lateral part
gives rise to hyperesthesia extending both above and below the lesion level on
the section side, while section of the whole Lissauers tract at any one level
had prolonged release effect on the next headward dermatome. Involvement
of Lissauers tract might justify the at-level hyperesthesia on the lesion side after
cord hemisection, Lissauers tract section, section of the posterior columns
impinging on the dorsal horn, and girdle pains in spinal tumors.
Roughly half or more of patients with syringomyelia suffer from a blend of
at-level and below-level pains. The onset of pain associated with traumatic
syringomyelia usually occurs in those dermatomes just above the level of the
trauma, but may sometimes be referred to distant dermatomes as the cyst encroaches
on high spinal cord segments.
In contrast to other types of pain that usually respond well to surgical treatment
of syrinx, dysesthetic CP can persist or even increase postoperatively, despite
collapse of the syrinx, and actually new CP can appear ex novo after surgical
treatment (Tator and Agbi 1991). In the series of Milhorat and colleagues (1996),
surgical treatment of syrinx resulted in total relief in only 7 of 37 patients (19%),

Piecing Together the Evidence

with other 15 improved of their dysesthetic pain; 15 patients (41%) reported no

improvement or even worsening of pain, despite MR-confirmed collapse of syrinx.
Postoperative dysesthetic pain was often disabling and poorly responsive to drugs.
One year after surgery, all these 15 patients continued to complain of dysesthesias
and pain, although at a lesser level in 9, and most even at 2
6 years postoperatively.
Syrinxes often encroached on the dorsolateral quadrant of the cord, but no
comparison between pain and non-pain patients was attempted in order to define a
possible role of the descending dorsolateral funiculus; similar arguments apply to
increase of substance P staining in the dorsal horns below-level and marked
reduction or absence at-level (references in Milhorat et al. 1997). However, Hida and
associates (1994) found that the syrinx cavity in posttraumatic syrinx patients was
more central at the caudal than at the rostral end. Sudden onset of pain immediately
above the original injury level is the most common presenting complaint from
patients with syrinx and often occurs in conjunction with a sudden increase in
thoracic pressure (e.g., during a sneeze). Milhorat and associates (1997) noted that
patients with syrinx pressures greater than 7.7 cmH2O tended to have more rapidly
progressive symptoms, exhibited greater improvement after shunting and a higher
incidence of postoperative dysesthetic pain than patients with normal or almost
normal pressures (30% versus 0%). Postoperative dysesthetic pain was not found
to be due to injury of dorsal roots or posterior columns during myelotomy and
chronic irritation of cord by shunt catheter, but only to sudden decompression of
hypertensive syrinxes. Such pains resolved spontaneously in two, were less severe
in another two, but persisted in a fifth at 1 year: these may have been segmental
dysesthetic pains, though. In 75% of patients with pre-drainage SSEPs abnormalities,
decompression produced a consistent reduction of N20 latencies and a similar,
but less consistent, increase in N20 amplitude. However, all comparisons between
high- and low-pressure groups were not statistically significant.
Attal and associates (2004) found that shunting of syrinx significantly improved
proprioceptive deficits, but not the magnitude of thermoalgesic deficits in
15 patients, despite collapse of the cavity in 80% of the cases: only pain evoked by
effort
cough
movement, but not pain at rest, was reduced at 2 years. Moreover,
only patients operated within less than 2 years of symptoms onset were improved or
stabilized, including 3 patients whose spontaneous pain improved by at least 70%.
Not finding a correlation between pain and thermoalgesic deficits, they suggested
that pain may result from irritation of the cord at the rostral end of the cyst. A similar
conclusion was reached in a study of subjects whose syrinxes were drained and
filled with fetal neural grafts (Wirth et al. 2002). Despite clear MRI evidence of
at least partial cyst obliteration in 7 subjects, complete disappearance of one or
more pain symptoms was noted only if collapse of the most rostral portion of the cyst
was achieved and no previous or new shunt tube was present in the cyst, suggesting
that syringomyelia pain may result from or be exacerbated by irritation of the cord levels
immediately rostral to the cyst. Irritation may be due to either a mass effect secondary
to increased cyst pressure and/or inflammation from tissue damage. In one patient,
reopening of a collapsed cyst seemed to cause return of pain. One subject noted
a delayed increase of pain after surgery, due to a delayed expansion of a second cyst

335

336

Central Pain Syndrome

distant from the transplant site. Pain intensity reports often varied substantially in
time, with distribution of dysesthesias more stable. However, complete disappearance
of a dysesthesia was seen in only 2 of 8 subjects. In one patient, the burning sensation
in the dermatomes associated with an upper C6
T3 cyst disappeared immediately
after grafting without shunting (follow-up 2 years), with complete collapse of cyst.
Nonetheless, he developed stabbing pain in the T6
9 dermatomes 3 months after
surgery due to expansion of the lower T6
9 cyst, both gradually increasing over
18 months; 27 months after the first surgery, a second graft was placed in this
lower cyst, with unsatisfactory results at 1 year, despite 50% collapse of the cyst.
Subject 5 had her previous stabbing pain in her legs limited to below knees at 6 weeks
and complete disappearance at 9 months (complete obliteration of cyst at 9 months),
but full relapse at 18 months (slight reopening at 12 months and persistence through
2 years). In the other 6 unrelieved patients, 5 had substantial collapse of the cyst
at the graft site, but also a persistent cyst above the graft site or shunt tubes at or
above the graft site. In the ninth, no collapse was seen.
However, Durward and colleagues (1982) reported that, although the syrinx
continued upward for many segments above the level of cordectomy and the upper
ends of the specimens of the cord showed pathological changes in three of their
patients, they were all relieved of their arm pain, indicating that this type of
abnormality may not be a generator of pain. On the other hand, in none of other
three cases where cordectomy failed was the rostral incision into histologically
normal cord. In two of them with a post-traumatic syringomyelia, earlier drainage
of the cyst had improved the syndrome, with the exception of the continuing pain.
The pain in these latter three failures was all referred well below the level of the lesion
in the cord, and these lesions were all at levels at which Jeffersons cordectomies had
also failed.
In patients with dysesthetic CP due to intramedullary tumors, symptoms tend
to persist in many after removal (Epstein et al. 1993; McCormick et al. 1990).
Surgical removal of intramedullary cavernomas may relieve CP initially, but many
relapse at follow-up (Kim et al. 2006). Also, new CP can appear after excision of
the mass (Canavero et al. 1994).
In sum, STT damage is the primary event in both BCP and CCP. While segmental
pains engage local processes, below-level diffuse pains, even in the best series, are
not uniformly relieved (unlike end-zone pains), so that we may conclude that cord
foci of hyperactivity play a boosting role only.
THE GENESIS OF ALLODYNIA
That allodynia is not pivotal to CP is proved by the simple observation that not all
patients complain of it, unlike steady spontaneous pain. As observed in exceptional
patients, allodynia may follow a different time course than spontaneous pain.
Greenspan and colleagues (1997) reported a woman with a thalamic lesion observed
over 4 years who had CP only during 3 months. Prior to spontaneous pain, there was
transient, but intense thermal allodynia several months before. Attal and colleagues
(1998) described a patient who presented uniquely with very intense brush-induced

Piecing Together the Evidence

allodynia (dynamic mechanical) strictly confined to the left C2/3 dermatomes

for several months. Thereafter, spontaneous pain and sensory deficits appeared
and a new MRI showed an intraspinal lesion involving the C2/5 segments.
Greenspan and associates (2004), on the basis of a study of 13 CPSP patients,
concluded that sparing of a submodality by lesions causing CP is associated with the
occurrence of allodynia in that modality, i.e., both tactile and cold/heat allodynia,
even striking, were significantly associated with the presence, rather than the absence
or reduction, of normal tactile and thermal sensibility. Similar observations have
been reported in syringomyelia (Ducreux et al. 2006). It was also noted how all
four patients with insular (posterior) lesions had tactile allodynia, but only one had
tactile sensory loss. However, both patients with insular lesions and noninsular
lesions had tactile allodynia, cold allodynia and thermotactile sensory deficits without
significant differences. Also on the basis of microstimulation studies (Chapter 7,
Section 2), Greenspan and associates have suggested that the termination of the
STT in the thalamus is reorganized to signal pain instead of cold in CP patients.
Cold allodynia would be due to input from an intact cold pathway driving Vc (and
not from loss of such input, disinhibiting these regions; see also Garcia-Larrea et al.
2002). Tactile allodynia would be due to disinhibition of Vc from loss of insula
or SI/SII input. In this context, Berics focus on a dissociation between STT and
(spared) DC-ML conduction would be redirected to explain tactile allodynia.
Tasker (2001, and references therein) observed how the induction of burning
and pain appears to be peculiar to patients with pain. Since all those in whom pain
was induced and half those in whom burning occurred suffered from evoked pain,
the phenomenon may be unrelated to the spontaneous pain (central allodynia).
He also noted that allodynia and hyperpathia in CPSP appear to be suppressed by
PVG DBS, as if depending on spinothalamic transmission. This central allodynia
occurs at sites where normally non-painful sensations are evoked, as well at sites
where normally no sensations are evoked, being unrelated spatially to the presence
of bursting or thalamic reorganization: he ascribed it to third-order neuron
sensitization. He also observed how evoked pains in SCI patients may be due
to conduction through spinothalamic pathways, and thus differing from steady
pain (Tasker et al. 1992). Sang and colleagues (1999) also concluded that SCI results
in central sensitization, accounting for cephalad spread of cold allodynia and of
augmented temporal summation.
Quantitative sensory studies and differential responses to drugs seem to indicate
that not all evoked pains have the same genesis, with a difference between
thermal evoked pain (amitriptyline responsive) and mechanical evoked pains
(lidocaine-morphine responsive). This would argue against a generalized hyperexcitability of nociceptive neurons to any type of stimuli (Attal et al. 2000, 2002).
Also, the effects of morphine on static mechanical allodynia suggest that static and
dynamic (brush evoked) mechano-allodynia associated with CP are sustained by
different mechanisms (brush-evoked allodynia having a similar genesis as in PNP).
In this regard, it should be noted that some opioids are weak NMDA, but not
AMPA, blockers: being hyperalgesia a supposedly NMDA-mediated phenomenon
(but see above), this might explain opioid action on hyperalgesia. However, our
own studies (Canavero and Bonicalzi 2004) show that both spontaneous pain

337

338

Central Pain Syndrome

and allodynia can be abolished simultaneously, although the latter to a greater

extent
or even exclusively
in some cases. GABA agonism may thus affect the
whole spectrum of CP.
Data suggest that some patients with cold allodynia tend to have more dorsally
placed thalamic lesions than those without, and those with movement allodynia
more anteriorly placed lesions (Bowsher 2005b).
Thus, sensitization at cord, brainstem and thalamic levels may play a role in the
genesis of allodynia only, but not spontaneous pain, with inappropriate activation
of the STT through stimulation of receptors and fibers that normally are not
involved in nociception. In other words, allodynia could be the result of exaggerated spinal input processed by an arrhythmic thalamus. As we have seen in reviewing
neurometabolic studies, allodynia is subserved by additional, widespread activity
particularly in frontal areas, perhaps justifying its high unpleasantness. However,
sudden disappearances as reviewed above strongly suggest that, once the loop
sustaining spontaneous pain has been switched off, allodynia is abolished
simultaneously.
Finally, there is no direct proof that expanded receptive fields play a role in human
patients.

REFERENCES

Ackermann RF, Finch DM, Babb TL, Engel J Jr.

(1984) Increased metabolism during longduration recurrent inhibition of hippocampal
pyramidal cells. J Neurosci 4, 251
264.
Ackerman LL, Follett KA, Rosenquist RW (2003)
Long-term outcomes during treatment of chronic
pain with intracisternal clonidine or clonidine/
opioid combinations. J Pain Symptom Manage
26, 668
677.
Agnew DA, Goldberg VD (1976) A brief trial of
phenytoin for thalamic pain. Bull Los Angeles
Neurol Soc 41, 9
12.
Ahn SH, Park HW, Lee BS, Moon HW, Jang SH,
Sakong J, Bae JH (2003) Gabapentin effect
on neuropathic pain compared among patients
with spinal cord injury and different durations
of symptoms. Spine 28, 341
346.
Akil H, Richardson DE, Hughes DE, Barchas JD
(1978) Enkephalin-like material elevated in
ventricular cerebrospinal fluid of pain patients
after analgesic focal stimulation. Science 201,
463
465.
Alajouanine T (1957) La douleur et les douleurs.
Paris: Masson.
Alajouanine T, Brunelli A (1935) Les douleurs
alternees dans les lesions bulbo-protuberantielles.
Contribution a` letude de la physiopathologie des
douleurs centrales. Revue Neurol 63, 828
837.
Albe-Fessard D (1973) Electrophysiological methods
for the identification of thalamic nuclei. Z Neurol
205, 15
28.
Albert ML (1969) Treatment of pain in multiple
sclerosis: preliminary report. N Engl J Med 280,
1395.
Amancio EJ, Peluso CM, Santos AC, Pena-Dias AP,
Debs FA (2002) [Central pain due to parietal
cortex compression by cerebral tumor: report of 2
cases.] Arq Neuropsiquiatr 60, 487
489.

Amano K et al. (1976) [in Japanese] cited from

Amano 1998.
Amano K, Iseki H, Notani M et al. (1980) Rostral
mesencephalic reticulotomy for pain relief. Report
of 15 cases. Acta Neurochir Suppl 30, 391
393.
Amano K, Tanikawa H, Hiroshi I et al. (1982)
Endorphins and pain relief. Further observations
on electrical stimulation of the lateral part of
the periaqueductal gray matter during rostral
mesencephalic reticulotomy for pain relief.
Appl Neurophysiol 45, 123
135.
Amano K, Kawamura H, Tanikawa T et al. (1986)
Long-term follow-up study of rostral mesencephalic reticulotomy for pain relief. Report of 34
cases. Appl Neurophysiol 49, 105
111.
Amano K, Kawamura H, Tanikawa T, Kawabatake
H, Iseki H, Taira T (1992) Stereotactic mesencephalotomy for pain relief. A plea for stereotactic
surgery. Stereotact Funct Neurosurg 59, 25
32.
Amano K (1998) Destructive central lesions for
persistent pain: II. Outcome. In Gildenberg PL,
Tasker RR, eds., Textbook of Stereotactic and
Functional Neurosurgery. New York: McGraw-Hill,
pp. 1425
1429.
Ameri D (1967) Sindrome talamico de Dejerine
Roussy en una meningitis tuberculosa de la
infancia. Arch Argent Pediatria 65, 173
178.
Andersen G, Vestergaard K, Lauritzen L (1994)
Effective treatment of post-stroke depression with
the selective serotonin re-uptake inhibitor citalopram. Stroke 25, 1099
1104.
Andersen G, Vestergaard K, Ingeman-Nielsen M,
Jensen TS (1995) Incidence of central post-stroke
pain. Pain 61, 187
193.
Anderson VC, Burchiel KJ (1999) A prospective
study of long-term intrathecal morphine in the
management of chronic nonmalignant pain.
Neurosurgery 44, 289
301.
339

340
Andy OJ (1983) Thalamic stimulation for chronic
pain. Appl Neurophysiol 46, 116
123.
Angel IF, Gould HJ Jr., Carey ME (1998) Intrathecal
morphine pump as a treatment option in chronic
pain of nonmalignant origin. Surg Neurol 49,
92
98.
Antonucci O (1938) Chordotomia posterior
medialis (dei cordoni di Goll) nelle paraplegie
spastiche (tipo Little). Policlinico (Sez Prat) 39,
1761
1768.
Apkarian AV, Stea RA, Mangios SH, Szeverenyi NM,
King RB, Thomas FD (1992) Persistent pain
inhibits controlateral somatosensory cortical
activity in humans. Neurosci Lett 140, 141
147.
Apkarian AV, Stea RA, Bolanowski SJ (1994)
Heat-induced pain diminishes vibrotactile perception: a touch gate. Somatosens Mot Res 11,
259
267.
Apkarian AV, Bushnell MC, Treede R-D, Zubieta J-K
(2005) Human brain mechanisms of pain perception and regulation in health and disease. Eur
J Pain 9, 463
484.
Armour D (1927) Lettsonian lecture on the surgery
of the spinal cord and its membranes. Lancet 1,
691
697.
Arner S, Meyerson BA (1988) Lack of analgesic effect
of opioids on neuropathic and idiopathic forms
of pain. Pain 33, 11
23.
Arseni G, Boetz MI (1971) Tulburari viscerovegetative si trofice in leziunile encefalice.
Bucharest: Editura Academiei Republicii Socialiste
Romania.
Ashby P, Rothwell JC (2000) Neurophysiologic
aspects of deep brain stimulation. Neurology
55(Suppl. 6), S17
S20.
Attal N, Brasseur B, Parker F, Chauvin M,
Bouhassira D (1998a) Effects of gabapentin on
the different components of peripheral and central
neuropathic pain syndromes: pathophysiological
considerations. Eur Neurol 40, 191
200.
Attal N, Brasseur L, Chauvin M, Bouhassira D
(1998b) A case of pure dynamic mechanoallodynia due to a lesion of the spinal cord:
pathophysiological considerations. Pain 75,
399
404.
Attal N, Gaude` V, Brasseur L et al. (2000)
Intravenous lidocaine in central pain. A doubleblind, placebo-controlled, psychophysical study.
Neurology 54, 564
574.
Attal N, Guirimand F, Brasseur L et al. (2002)
Effects of IV morphine in central pain. A randomized placebo-controlled study. Neurology 58,
554
563.

References
Attal N, Brasseur L, Guirimand D et al. (2004) Are
oral cannabinoids safe and effective in refractory
neuropathic pain? Eur J Pain 8, 173
177.
Attal N, Parker F, Tadie` M, Bouhassira D (2004)
Effects of surgery on central pain and sensory
deficits associated with syringomyelia: a longterm prospective psychophysical study. J Neurol
Neurosurg Psych 75, 1025
1030.
Awerbuch GI, Sandyk R (1990) Mexiletine for
thalamic pain syndrome. Int J Neurosci 55,
129
133.
Babtchine IS (1936) Les resultats immediats et
lontains de la cordotomie. J Chir (Par) 47, 26
39.
Backonja M, Gombar K (1992) Response of central
pain syndromes to intravenous lidocaine. J Pain
Symptom Manage 7, 172
178.
Backonja M, Howland EW, Wang J et al. (1991)
Tonic changes in alpha power during immersion
of the hand in cold water. Electroenceph Clin
Neurophysiol 79, 192
203.
Backonja M, Arndt G, Gombar KA, Check B,
Zimmermann M (1994) Response of chronic
neuropathic pain syndromes to ketamine: a
preliminary study. Pain 56, 51
57.
Bailey RA, Glees P, Oppenheimer DR (1954)
Midbrain tractotomy. A surgical and clinical
report, with observations on ascending and descending tract degeneration. Mschr Psychiat Neurol
127, 316
335.
Bainton T, Fox M, Bowsher D, Wells C (1992) A
double-blind trial of naloxone in central poststroke pain. Pain 48, 159
162.
Ballantine HT Jr., Giriunas IE (1988) Treatment of
intractable psychiatric illness and chronic pain by
stereotactic cingulotomy. In Schmideck W, Sweet
WH, eds., Operative Neurosurgical Techniques, 2nd
edn. New York: Grune-Stratton, pp. 1069
1075.
Ballantyne JC, Mao J (2003) Opioid therapy for
chronic pain. New Engl J Med 349, 1943
1953.
Banerjee T (1974) Transcutaneous nerve stimulation
for pain after spinal injury. New Engl J Med
291, 796.
Bang OY, Lee JS, Lee PH, Lee G (2005) Autologous
mesenchymal stem cell transplantation in stroke
patients. Ann Neurol 57, 874
882.
Baranauskas G, Nistri A (1998) Sensitization of pain
pathways in the spinal cord: cellular mechanisms.
Progr Neurobiol 54, 349
365.
Barcia-Salorio JL, Roldan P, Lopez-Gomez L (1987)
Radiosurgery of central pain. Acta Neurochir
(Wien) Suppl 39, 159
162.
Barolat G (1995) Current status of epidural spinal
cord stimulation. Neurosurg Quart 5, 98
124.

References
Barolat G, Ketcik B, He J (1998) Long-term outcome
of spinal cord stimulation for chronic pain
management. Neuromodulation 1, 19
25.
Baron JC, DAntona R, Pantano R et al. (1986)
Effects of thalamic stroke on energy metabolism
of the cerebral cortex. Brain 109, 1243
1259.
Baron R, Baron Y, Disbrow E, Roberts TP (1999)
Brain processing of capsaicin-induced secondary
hyperalgesia. A functional MRI study. Neurology
53, 548
557.
Barraquer-Bordas L, Molet J, Pascual-sedano B,
Catala` H (1999) Dolor central retardado asociado
a hematoma subinsular seguido por tumor parietooccipital. Efecto favorable de la estimulacion
cronica del nucleo VPL talamico. Rev Neurologia
29, 1044
1048.
Bassetti C, Bogousslavsky J, Mattle H, Bernasconi A
(1997) Medial medullary stroke: report of seven
patients and review of the literature. Neurology
48, 882
890.
Bassetti RD, Bogousslavsky J, Regli F (1993) Sensory
syndromes in parietal stroke. Neurology 43,
1942
1949.
Bates J, Nathan PW (1980) Transcutaneous electrical
nerve stimulation for chronic pain. Anesthesia 35,
817
822.
Baudoin A, Puech P (1949) Premiers essais dintervention directe sur le thalamus (injection, electrocoagulation). Rev Neurol 81, 78
81.
Beatty RA (1970) Cold dysesthesia. A symptom
of extramedullary tumors of the spinal cord.
J Neurosurg 33, 75
78.
Becker R, Uhle EI, Alberti O, Bertalanffy H (2000)
Continuous intrathecal baclofen infusion in the
management of central deafferentation pain.
J Pain Symptom Manage 20, 313
315.
Behan RJ (1914) Pain: Its Origin, Conduction,
Perception and Diagnostic Significance. New York:
Appleton.
Bejjani B-P, Arnulf I, Houeto J-L et al. (2002)
Concurrent excitatory and inhibitory effects of
high frequency stimulation: an oculomotor study.
J Neurol Neurosurg Psych 72, 517
522.
Bejjani GK, Cockerham KP (2001) Adult Chiari
malformation. Contemp Neurosurg 23.
Belfrage M, Segerdahl M, Arner S, Sollevi A (1999)
The safety and efficacy of intrathecal adenosine
in patients with chronic neuropathic pain. Anesth
Analg 89, 136
142.
Bell E, Karnosh LJ (1949) Cerebral hemispherectomy. J Neurosurg 6, 285
293.
Bender MB, Jaffe R (1958) Pain of central origin.
Med Clin North Am 49, 691
700.

341
Bendok B, Levy RM (1998) Brain stimulation for
persistent pain management. In Gildenberg PL,
Tasker RR, eds., Textbook of Stereotactic and
Functional Neurosurgery. New York: McGraw-Hill,
pp. 1539
1546.
Berglund B, Harju E-L, Lindblom U (2001) Central
and peripheral neuropathic pain characterized
by perceived intensity and quality of touch, cold,
and warmth. Arch Center Sensory Res 6, 31
53.
Beric A, Dimitrijevic MR, Lindblom U (1988)
Central dysesthesia syndrome in spinal cord injury
patients. Pain 34, 109
116.
Beric A (1993) Central pain: new syndromes and
their evaluation. Muscle Nerve 16, 1017
1024.
Beric A (1999) Spinal cord damage: injury. In Wall
PD, Melzack R, eds., Textbook of Pain, 4th edn.
Edinburgh: Churchill Livingstone, pp. 915
927.
Berthier M, Starkstein S, Leiguarda R (1988)
Asymbolia for pain: a sensory-limbic disconnection syndrome. Ann Neurol 24, 41
49.
Bettag W (1966) Results of treatment of pain
by interruption of the medial pain tract of the
brain stem. Excerpta Med Int Cong Ser 110,
771
775.
ber stereotaktischen
Bettag W, Yoshida T (1960) U
schmerzoperationen. Acta Neurochir 8, 299
317.
Biella G, Salvadori G, Sotgiu ML (1999) Multifractal
analysis of wide dynamic range neuron discharge
profiles in normal rats and in rats with sciatic
nerve constriction. Somatosens Motor Res 16,
89
102.
Biemond A (1956) The conduction of pain above
the level of the thalamus opticus. Arch Neurol
Psych 75, 231
244.
Binder A, Schattschneider J, Wolff S et al. (2002)
Inhibition of human motor cortex by tonic
cutaneous pain. A fMRI study. In Abstracts, 10th
World Congress on Pain. Seattle, WA: IASP Press,
Abst. 1125-P41.
Bittar RG, Ptito A, Reutens DC (2000) Somatosensory representation in patients who have undergone hemispherectomy: a functional magnetic
resonance imaging study. J Neurosurg 92, 45
51.
Blond S, Touzet G, Reyns N et al. (2000) Les
techniques de neurostimulation dans le traitement
de la douleur chronique. Neurochirurgie 46,
466
482.
Blumberg H (1988) Zur Entstehung und Therapie
des Schmerzsyndroms bei der sympathischen
Reflexdystrophie. Der Schmerz 2, 125
143.
Boas RA, Corvino BG, Shahnarian A (1982)
Analgesic responses to i.v. lignocaine. Br J Anaesth
54, 501
505.

342
Bogousslavsky J, Regli F, Uske A (1988) Thalamic
infarcts: clinical syndromes, etiology, and
prognosis. Neurology 38, 837
48. Erratum in
Neurology 1988, 38, 1335.
Bohm E (1960) Chordotomy for intractable pain due
to malignant disease. Acta Psych Neurol Scand 35,
145
155.
Boivie J, Leijon G, Johansson I (1989) Central
post-stroke pain. A study of the mechanisms
through analyses of the sensory abnormalities.
Pain 37, 173
185.
Boivie J, Leijon G (1991) Clinical findings in patients
with central poststroke pain. In Casey KL, ed.,
Pain and Central Nervous System Disease. The
Central Pain Syndromes. New York: Raven Press,
pp. 65
75.
Bonica JJ (1953) The Management of Pain.
Philadelphia, PA: Lea and Febiger.
Bonica JJ (1991) Introduction: semantic, epidemiologic, and educational issues. In Casey KL, ed.,
Pain and Central Nervous System Disease.
New York: Raven Press, pp. 13
30.
Bonicalzi V, Canavero S, Cerutti F et al. (1997)
Lamotrigine reduces total postoperative analgesic requirement: a randomized double-blind,
placebo-controlled pilot study. Surgery 122,
567
570.
Bonicalzi V, Canavero S (1999a) CRPS: are guidelines possible? Clin J Pain 15, 159
169.
Bonicalzi V, Canavero S (1999b) Comments on
Kingery. Pain 73, (1997) 123
139. Pain 79,
317
323.
Bonicalzi V, Canavero S (2000) Sympathetic pain
again? Lancet 360, 1426
1427.
Bonicalzi V, Canavero S (2004) Intrathecal ziconotide for chronic pain. JAMA 292, 1681
1682.
Bornstein B (1949) Sur le phenome`ne du membre
fantome. Encephale 38, 32
46.
Bors E (1951) Phantom limbs of patients with spinal
cord injury. Arch Neurol Psych 66, 610
631.
Bosch DA (1991) Stereotactic rostral mesencephalotomy in cancer pain and deafferentation pain.
A series of 40 cases with follow-up results.
J Neurosurg 75, 747
751.
Botterell EH, Callaghan JC, Jousse T (1954) Pain
in paraplegia. Clinical management and surgical
treatment. Proc R Soc Med 47, 281
288.
Bouchard G, Mayanagi Y, Martins LF (1977)
Advantages and limits of intracerebral stereotactic operations for pain. In Sweet DH,
Obradors, Martin-Rodriguez JC, eds., Neurosurgical Treatment in Psychiatry, Pain and Epilepsy. Baltimore, MD: University Park Press,
pp. 693
696

References
Bouckoms AF (1989) Psychosurgery for pain.
In Wall PD, Melzack R, eds., Textbook of Pain.
Edinburgh: Churchill Livingstone, pp. 868
881.
Bouhassira D, Attal N, Brasseur L, Parker F (2000)
Quantitative sensory testing in patients with
painful or painless syringomyelia. In Devor M,
Rowbotham MC, Wiesenfeld-Hallin Z, eds.,
Proceedings of the 9th World Congress on Pain.
Seattle, WA: IASP Press, pp. 401
409.
Bowsher D, Lahuerta J (1987) A case of tabes dorsalis
with tonic pupils and lightning pains relieved by
sodium valproate. J Neurol Neurosurg Psychiatry
50, 239
241.
Bowsher D, Foy PM, Shaw MDM (1989) Central
pain complicating infarction following subarachnoid hemorrhage. Br J Neurosurg 3, 435
442.
Bowsher D (1989) Contralateral mirror-image pain
following anterolateral cordotomy. Pain 33,
63
65.
Bowsher D (1993) Sensory consequences of stroke.
Lancet 34, 156.
Bowsher D (1994) La douleur neurogene. Doul et
Analg 2, 65
69.
Bowsher D (1995) The management of central poststroke pain. Postgrad Med J 71, 599
604.
Bowsher D (1996) Central pain: clinical and
physiological characteristics. J Neurol Neurosurg
Psych 61, 62
69.
Bowsher D, Leijon G, Thomas KA (1998) Central
poststroke pain: correlation of MRI with clinical
pain characteristics and sensory abnormalities.
Neurology 51, 1352
1358.
Bowsher D (2001) Stroke and central poststroke
pain in an elderly population. J Pain 2, 258
261.
Bowsher D (2002) The time factor in tricyclic
treatment of postherpetic neuralgia (PHN) and
central post-stroke pain (CPSP). In 10th World
Congress on Pain, Book of Abstracts. Seattle, WA:
IASP Press, A890
P160.
Bowsher D (2005a) Dynamic mechanical allodynia
in neuropathic pain. Pain 116, 164
165.
Bowsher D (2005b) Allodynia in relation to lesion
site in central post-stroke pain. J Pain 6, 736
740.
Bowsher D, Haggett C (2005) Paradoxical burning
sensation produced by cold stimulation in patients
with neuropathic pain. Pain 117, 230.
Breuer A, Cuervo H, Selkoe DJ (1981) Hyperpathia
and sensory level due to parietal lobe arteriovenous malformation. Arch Neurol 38, 722
724.
Brihaye J, Retif J (1961) Comparaison des resultats
obtenus par la cordotomie antero-laterale au
niveau dorsal et au niveau cervical. A propos de
109 observations personnelles. Neurochirurgie 7,
258
277.

References
Broager B (1974) Commissural myelotomy. Surg
Neurol 2, 71
74.
Broggi G, Franzini A, Giorgi C, Servello D,
Spreafico R (1984) Preliminary results of specific
thalamic stimulation for deafferentation pain.
Acta Neurochir Suppl 33, 497
500.
Broggi G, Servello D, Dones I, Carbone G (1994)
Italian multicentric study on pain treatment with
epidural spinal cord stimulation (SCS). Stereotact
Funct Neurosurg 62, 273
278.
Brooks J, Nurmikko TJ, Bimson B (2002) FMRI
studies of thermosensation and nociception using
graded thermal stimuli. In 10th World Congress on
Pain, Book of Abstracts. Seattle, WA: IASP Press,
A1113
P29.
Brooks-Kayal AR, Shumate MD, Jin H, Rikhter TY,
Coulter DA (1998) Selective changes in single
cell GABA(A) receptor subunit expression and
function in temporal lobe epilepsy. Nature
Med 4, 1166
72. Erratum in: Nature Med 1999,
5, 590.
Browder J, Gallagher JP (1948) Dorsal cordotomy
for painful phantom limbs. Ann Surg 128,
456
469.
Brown JA, Pilitsis JG (2005) Motor cortex stimulation for central and neuropathic facial pain: a
prospective study of 10 patients and observations
of enhanced sensory and motor function during
stimulation. Neurosurgery 56, 290
297.
Buchanan M (2002) Nexus. Small Worlds and the
Groundbreaking Science of Networks. WW Norton.
Buchhaas U, Koulousakis A, Nittner K (1989)
Experience with spinal cord stimulation (SCS) in
the management of chronic pain in a traumatic
transverse lesion syndrome. Neurosurg Rev
12(Suppl. 1), 582
587.
Budd K (1985) The use of the opiate antagonist
naloxone in the treatment of intractable pain.
Neuropeptides 5, 419
422.
Bullit E (1991) Abnormal anatomy of deafferentation: regeneration and sprouting within the
central nervous system. Adv Pain Res Ther 19.
Burchiel KJ, Russel LC (1987) Has the amount of
spontaneous electrical activity in experimental
neuromas been overestimated? In Pubols LM,
Sessle BJ, eds., Effects of Injury on Trigeminal and
Spinal Somatosensory Systems. New York: Liss,
pp. 77
83.
Burke DC (1973) Pain in paraplegia. Paraplegia 10,
297
313.
Bushnell MC, Duncan GH, Hofbauer RK et al.
(1999) Pain perception: is there a role for primary
somatosensory cortex? Proc Natl Acad Sci USA 96,
7705
7709.

343
Cahana A, Carota A, Montadon ML, Annoni JM
(2004) The long-term effect of repeated intravenous lidocaine on central pain and possible
correlation in positron emission tomography
measurements. Anesth Analg 98, 1581
1584.
Calford MB, Tweedale R (1990) Interhemispheric
transfer of plasticity in the cerebral cortex. Science
249, 805
807.
Cameron T (2004) Safety and efficacy of spinal
cord stimulation for the treatment of chronic
pain: a 20-year literature review. J Neurosurg 100
(Suppl. 3), 254
267.
Campanini A, De Risio C (1962) Sul problema del
dolore di origine centrale. Sist Nerv 14, 382
386.
Canavero S, Pagni CA, Castellano G et al. (1993)
The role of cortex in central pain syndromes:
preliminary results of a long-term technetium-99
hexamethylpropyleneamineoxime single photon
emission computed tomography study. Neurosurgery 32, 185
191.
Canavero S (1994) Dynamic reverberation. A unified
mechanism for central and phantom pain.
Med Hypotheses 42, 203
207.
Canavero S, Pagni CA, Duca S, Bradac GB (1994)
Spinal intramedullary cavernous angiomas: a
literature metaanalysis. Surg Neurol 41, 381
388.
Canavero S, Bonicalzi V, Pagni CA et al. (1995a)
Propofol analgesia in central pain: preliminary
clinical observations. J Neurol 242, 561
567.
Canavero S, Pagni CA, Bonicalzi V (1995b) Transient hyperacute allodynia in Schneiders syndrome: an irritative genesis? Ital J Neurol Sci 16,
555
557.
Canavero S, Bonicalzi V (1995) Cortical stimulation
for central pain. J Neurosurg 83, 1117.
Canavero S, Bonicalzi V, Castellano G (1996) Two in
one: the genesis of central pain. Pain 64, 394
395.
Canavero S (1996) Bilateral central pain. Acta Neurol
Belg 96, 135
136.
Canavero S, Bonicalzi V (1996) Lamotrigine control
of central pain. Pain 68, 179
181.
Canavero S, Bonicalzi V, Massa-Micon B (1997)
Central neurogenic pruritus: a literature review.
Acta Neurol Belg 97, 244
247.
Canavero S, Bonicalzi V (1998a) Review article. The
neurochemistry of central pain: evidence from
clinical studies, hypothesis and therapeutic implications. Pain 74, 109
114.
Canavero S, Bonicalzi V (1998b) Pain after thalamic
stroke: right diencephalic predominance and
clinical features in 180 patients. Neurology 51,
927
928.
Canavero S, Bonicalzi V, Castellano G, Perozzo P,
Massa-Micon B (1999) Painful supernumerary

344
phantom arm following motor cortex stimulation for central post-stroke pain. J Neurosurg
91, 121
123.
Canavero S, Bonicalzi V, Lacerenza M et al. (2001)
Disappearance of central pain following iatrogenic
stroke. Acta Neurol Belg 101, 221
223.
Canavero S, Bonicalzi V (2001a) Reversible central
pain. Neurol Sci 22, 271
273.
Canavero S, Bonicalzi V (2001b) Electroconvulsive
therapy and pain. Pain 89, 301
302.
Canavero S, Bonicalzi V, Paolotti R (2002a) Lack of
effect of topiramate for central pain. Neurology 58,
831
832.
Canavero S, Bonicalzi V, Paolotti R (2002b)
Reboxetine for central pain: a single-blind
prospective study. Clin Neuropharmacol 25,
238
239.
Canavero S, Bonicalzi V (2002) Therapeutic extradural cortical stimulation for central and neuropathic pain: a review. Clin J Pain 18, 48
55.
Canavero S, Bonicalzi V (2003a) Chronic neuropathic pain. New Engl J Med 348, 2688
2689.
Canavero S, Bonicalzi V (2003b) Neuromodulation
for central pain. Expert Rev Neurotherapeutics 3,
591
607.
Canavero S, Bonicalzi V, Dotta M, Vighetti S,
Asteggiano G (2003a) Low-rate repetitive TMS
allays central pain. Neurol Res 25, 151
152.
Canavero S, Bonicalzi V, Narcisi P (2003b) Safety
of magnesium-lidocaine combination for severe
head injury: the Turin Lidomag pilot study. Surg
Neurol 60, 165
169.
Canavero S, Bonicalzi V (2004a) Intravenous subhypnotic propofol in central pain. A double-blind,
placebo-controlled, crossover study. Clin Neuropharmacol 27, 182
186.
Canavero S, Bonicalzi V (2004b) Motor cortex
stimulation for central and neuropathic pain. Pain
108, 199
200.
Canavero S, Bonicalzi V (2005a) Transcranial
magnetic stimulation for central pain. Curr Pain
Headache Rep 9, 87
89.
Canavero S, Bonicalzi V (2005b) Mexiletinegabapentin for central pain: an efficacy and
long-term study. In 11th World Congress on
Pain, Book of Abstracts. Seattle, WA: IASP Press.
P218 A596 P202.
Canavero S, Bonicalzi V (2006) Extradural cortical
stimulation for central pain. In Sakas DE, Simpson
B, Krames E, eds., Operative Neuromodulation.
Vienna: Springer Verlag (in press).
Cannon WB, Rosenblueth A (1949) The Supersensitivity of Denervated Structures, a Law of
Denervation. New York: Macmillan.

References
Cantor F (1972) Phenytoin treatment of thalamic
pain. Br Med J 2, 590.
Cardenas DD, Warms CA, Turner JA et al. (2002)
Efficacy of amitriptyline for relief of pain in spinal
cord injury: results of a randomised controlled
trial. Pain 96, 365
373.
Carlsson KC, Hoem NO, Moberg ER, Mathisen LC
(2004) Analgesic effect of dextromethorphan in
neuropathic pain. Acta Anaesthesiol Scand 48,
328
336.
Carrieri PB, Provitera VV, Lavorgna L, Bruno R
(1998) Response of thalamic pain syndrome to
lamotrigine. Eur J Neurol 5, 625
626.
Carter ML (2004) Spinal cord stimulation in chronic
pain: a review of evidence. Anaesth Intensive Care
32, 11
21.
Casey KL, ed. (1991) Pain and Central Nervous
System Disease. The Central Pain Syndromes.
New York: Raven Press.
Casey KL, Beydoun A, Boivie J et al. (1996) Laserevoked cerebral potentials and sensory function in
patients with central pain. Pain 64, 485
491.
Casey KL, Morrow TJ, Lorenz J, Minoshima S (2001)
Temporal and spatial dynamics of human forebrain activity during heat pain: analysis by PET.
J Neurophysiol 85, 951
959.
Cassinari V, Pagni CA, Infuso L, Marossero F (1964)
La chirurgia stereotassica dei dolori incoercibili
(esperienza personale a proposito di 20 casi). Sist
Nerv 16, 17
28.
Cassinari V, Pagni CA (1969) Central Pain: A
Neurosurgical Survey. Cambridge, MA: Harvard
University Press, pp. 1
192.
Castro-Alamancos MA, Connors BW (1997)
Thalamocortical synapses. Progr Neurobiol 51,
581
606.
Cepeda MS, Africano JM, Polo R, Alcala R, Carr DB
(2003) What decline in pain intensity is meaningful
to patients with acute pain? Pain 105, 151
157.
Cesaro P, Mann MW, Moretti JL et al. (1991)
Central pain and thalamic hyperactivity: a single
photon emission computerized tomographic
study. Pain 47, 329
336.
Chatterjee A, Almahrezi A, Ware M, Fitzcharles MA
(2002) A dramatic response to inhaled cannabis in
a woman with central thalamic pain and dystonia.
J Pain Symptom Manage 24, 4
6.
Chen B, Stitik TP, Foyc PM, Nadler SF, DeLisa JA
(2002) Central post-stroke pain syndrome: yet
another use for gabapentin? Am J Phys Med Rehab
81, 718
720.
Chiou-Tan FY, Tuel SM, Johnson JC et al. (1996)
Effect of mexiletine on spinal cord injury dysesthetic pain. Am J Phys Med Rehab 75, 84
87.

References
Chung CS, Caplan LR, Han W et al. (1996) Thalamic
haemorrhage. Brain 119, 1873
1886.
Cianchetti C, Zuddas A, Randazzo AP, Perra L,
Marrosu MG (1999) Lamotrigine adjunctive
therapy in painful phenomena in MS: preliminary
observations. Neurology 53, 433.
Cioni B, Meglio M, Pentimalli L, Visocchi M (1995)
Spinal cord stimulation in the treatment of
paraplegic pain. J Neurosurg 82, 35
39.
Cioni B et al. (1996) cited from Canavero S,
Bonicalzi V (2002).
Clifford DB, Trotter JL (1984) Pain in multiple
sclerosis. Arch Neurol 41, 1270
1272.
Coghill RC, Sang CN, Maisog JM, Iadarola MJ
(1999) Pain intensity processing within the human
brain: a bilateral, distributed mechanism.
J Neurophysiol 82, 1934
1943.
Coghill RC, McHaffie JG, Yen YF (2003) Neural
correlates of interindividual differences in the
subjective experience of pain. Proc Natl Acad Sci
USA 100, 8538
8542.
Cohen SP, Abdi J (2002) Venous malformations
associated with central pain: report of a case.
Anesth Analg 95, 1358
1360.
Cohen SP, DeJesus M (2004) Ketamine patientcontrolled analgesia for dysesthetic central pain.
Spinal Cord 42, 425
428.
Cole JD, Illis LS, Sedgwick EM (1987) Pain produced by spinal cord stimulation in a patient with
allodynia and pseudo-tabes. J Neurol Neurosurg
Psych 50, 1083
1084.
Cole JD, Illis LS, Sedgwick EM (1991) Intractable
central pain in spinal cord injury is not relieved
by spinal cord stimulation. Paraplegia 29,
167
172.
Condouris GA (1976) Local anesthetics as modulators of neural information. In Bonica JJ,
Albe-Fessard DG, eds., Advances in Pain Research
and Therapy, vol. 1. New York: Raven Press,
pp. 663
667.
Constans JP (1960) Chirurgie frontale de la douleur.
Acta Neurochir(Wien) 8, 251
281.
Cook AW, Kawakami Y (1977) Commissural
myelotomy. J Neurosurg 47, 1
6.
Cook AW, Nathan PW, Smith MC (1984) Sensory
consequences of commissural myelotomy: a
challenge to traditional anatomical concepts.
Brain 107, 547
568.
Cooper IS (1965) Clinical and physiologic implications of thalamic surgery for disorders of sensory
communication: 1. Thalamic surgery for intractable pain. J Neurol Sci 2, 493
519.
Cooper IS, Amin I, Candra R, Waltz JM (1973) A
surgical investigation of the clinical physiology of

345
the LP-pulvinar complex in man. J Neurol Sci 18,
89
110.
Cossart R, Bernard C, Ben-Ari Y (2005) Multiple
facets of GABAergic neurons and synapses: multiple fates of GABA signalling in epilepsies. TINS 28,
108
113.
Cowie RA, Hitchcoch ER (1982) The late results of
antero-lateral cordotomy for pain relief. Acta
Neurochir 64, 39
50.
Crabtree JW, Collingridge GL, Isaac JT (1998) A new
intrathalamic pathway linking modality-related
nuclei in the dorsal thalamus. Nature Neurosci 1,
389
394.
Craig AD (1998) A new version of the thalamic
disinhibition hypothesis of central pain. Pain
Forum 7, 1
14.
Crawford AS, Knighton RS (1953) Further observations on medullary spino-thalamic tractotomy.
J Neurosurg 10, 113
121.
Crawford AS (1960) Medullary spinothalamic tractotomy for high intractable pain. J Maine Med
Assoc 51, 233
235.
Crisologo PA, Neal B, Brown R, McDanal J, Kissin I
(1991) Lidocaine-induced spinal block can
relieve central poststroke pain: role of the block
in chronic pain diagnosis. Anesthesiology 74,
184
185.
DAleo G, Sessa E, Di Bella P et al. (2001)
Topiramate modulation of R3 nociceptive reflex
in multiple sclerosis patients suffering paroxysmal
symptoms. J Neurol 248, 996
999.
Dandy et al. (1933) cited from Muller et al. (1991).
Daniele O, Fierro B, Brighina F, Magaudda A,
Natale` E (2003) Disappearance of haemorrhagic
stroke-induced thalamic (central) pain following
a further (contralateral ischaemic) stroke. Funct
Neurol 18, 95
96.
Dario A et al. (1999) cited from Canavero S,
Bonicalzi V (2002).
David M, Talairach J, Hecaen H (1947) Etude
critique des interventions neurochirurgicales
actuellement pratiquees dans le traitement de la
douleur. Semaine des Hopitaux de Paris 234,
1651
1665.
Davidoff G, Roth EJ (1991) Clinical characteristics of
central (dysesthetic) pain in spinal cord injury
patients. In Casey KL, ed., Pain and Central
Nervous System Disease. New York: Raven Press,
pp. 77
83.
Davidoff G, Guarracini M, Roth E, Sliwa J, Yarkony
G (1987a) Trazodone hydrochloride in the treatment of dysesthetic pain in traumatic myelopathy:
a randomized, double-blind, placebo-controlled
study. Pain 29, 151
161.

346
Davidoff G, Roth E, Guarracini M, Sliwa J,
Yarkony G (1987b) Function-limiting dysesthetic
pain syndrome among traumatic spinal cord
injury patients: a cross-sectional study. Pain 29,
39
48.
Davis KD, Hutchison WD, Lozano AM, Dostrovsky
JO (1994) Altered pain and temperature perception following cingulotomy and capsulotomy in
a patient with schizoaffective disorder. Pain 59,
189
199.
Davis KD, Kiss ZHT, Tasker RR, Dostrovsky JO
(1996) Thalamic stimulation-evoked sensations in
chronic pain patients and in nonpain (movement
disorder) patients. J Neurophysiol 75, 1026
1037.
Davis KD, Kiss ZH, Luo L et al. (1998) Phantom
sensations generated by thalamic microstimulation. Nature 391, 385
387.
Davis KD, Taub E, Duffner F et al. (2000) Activation
of the anterior cingulate cortex by thalamic
stimulation in patients with chronic pain: a
positron emission tomography study. J Neurosurg
92, 64
69.
Davis L, Martin J (1947) Studies upon spinal cord
injuries: II. The nature and treatment of pain.
J Neurosurg 4, 483
491.
Davis L (1954) Treatment of spinal cord injuries.
AMA Arch Surg 69, 488
495.
Davis R, Lentini R (1975) Transcutaneous nerve
stimulation for treatment of pain in patients
with spinal cord injury. Surg Neurol 4, 100
101.
Davis RA, Stokes JW (1966) Neurosurgical attempts
to relieve thalamic pain. Surg Gynecol Obster 123,
371
384.
Davison C, Schick W (1935) Spontaneous pain and
other subjective sensory disturbances. A clinicopathologic study. Arch Neur Psych 34, 1204
1237.
De Ajuriaguerra J (1937) La douleur dans les
affections du systeme nerveux central. Paris: Doin.
DeCharms RC, Maeda F, Glover GH et al. (2005)
Control over brain activation and pain learned by
using real-time functional MRI. Proc Natl Acad Sci
USA 102, 18626
18631.
Deckers CLP, Czuczwar SJ, Hekster YA et al. (2000)
Selection of antiepileptic drug polytherapy based
on mechanisms of action: the evidence reviewed.
Epilepsia 41, 1364
1374.
DeFelipe J, Farinas I (1992) The pyramidal neuron of
the cerebral cortex: morphological and chemical
characteristics of the synaptic inputs. Progr
Neurobiol 39, 563
607.
Defrin R, Ohry A, Blumen N, Urca G (2001)
Characterization of chronic pain and somatosensory function in spinal cord injury subjects. Pain
89, 253
263.

References
Dehen H, Willer JC, Cambier J (1983) Pain in
thalamic syndrome: electrophysiological findings
in man. Adv Pain Res Ther 5, 936
940.
Dejerine J, Egger M (1903) Contribution a` letude de
la physiologie pathologique de lincoordination
motrice. Rev Neurol 11, 397.
Dejerine J, Roussy G (1906) Le syndrome thalamique. Rev Neurol 14, 521
532.
Dellemijn PL, Vanneste JA (1997) Randomised
double-blind active-placebo-controlled crossover
trial of intravenous fentanyl in neuropathic pain.
Lancet 349, 753
758.
Demirel T, Braun W, Reimers CD (1984) Results of
spinal cord stimulation in patients suffering from
chronic pain after a two year observation period.
Neurochirurgia (Stuttg) 27, 47
50.
Denny-Brown D, Kirk EJ, Yanagisawa N (1973)
The tract of Lissauer in relation to sensory transmission in the dorsal horn of the spinal cord
in the macaque monkey. J Comp Neurol 151,
175
200.
Derbyshire SW, Jones AK, Gyulai F et al. (1997) Pain
processing during three levels of noxious stimulation produces differential patterns of central
activity. Pain 73, 431
445.
De Salles AF, Bittar GT Jr. (1994) Thalamic pain
syndrome: anatomic and metabolic correlations.
Surg Neurol 41, 147
151.
Devor M, Basbaum AL, Bennett GJ et al. (1991)
Group report: mechanisms of neuropathic pain
following peripheral injury. In Basbaum I, Besson
JM, eds., Towards a New Pharmacotherapy of
Pain. Chichester: Wiley, pp. 417
440.
Dey DD, Landrum O, Oaklander AL (2005) Central
neuropathic itch from spinal-cord cavernous
hemangioma: a human case, a possible animal
model, and hypotheses about pathogenesis.
Pain 113, 233
237.
Diamond ME, Huang W, Ebner FF (1994) Laminar
comparison of somatosensory cortical plasticity.
Science 265, 1885
1888.
Diaz JM, Slagle DC (1992) Abscesses and thalamic
pain. Neurology 42, 2306
2307.
Di Biagio F (1959) Dolore centrale da lesione
sopratalamica regredito con atophanyl. Riv Neurol
29, 476
481.
Dieckmann G, Witzmann A (1982) Initial and longterm results of deep brain stimulation for chronic
intractable pain. Appl Neurophysiol 45, 167
172.
Dieckmann G, Veras G II (1984) Plexus avulsion
pain (neurogenic pain). High frequency coagulation of the dorsal root entry zone in patients with
deafferentation pain. Acta Neurochir (Wien) Suppl
33, 445
450.

References
Diemath EE, Heppner F, Walker AE (1961) Anterolateral chordotomy for relief of pain. Postgrad
Med 29, 485
495.
Dierssen G, Odoriz B, Hernando C (1969) Sensory
and motor response to stimulation of the posterior cingulate cortex in man. J Neurosurg 31,
435
440.
Dimitrijevic MR (1987) Neurophysiology in spinal
cord injury. Paraplegia 25, 205
208.
Di Piero V, Jones AKP, Iannotti F et al. (1991)
Chronic pain: a PET study of the central effects of
percutaneous high cervical cordotomy. Pain 46,
9
12.
Djaldetti R, Shifrin A, Rogowski Z et al. (2004)
Quantitative measurement of pain sensation in
patients with Parkinson disease. Neurology 62,
2171
2175.
Dogliotti AM (1937) Trattamento del dolore nei
tumori, Minerva Med 28, 455
461.
Dogliotti AM (1938) First surgical sections in man,
of the lemniscus lateralis (pain-temperature paths)
at the brainstem, for the treatment of diffused
rebellious pain. Anesth Analg 17, 143
145.
Doi N, Nakamura M, Isse K et al. (1999) Electroconvulsive therapy for central post-stroke pain. In
9th World Congress on Pain, Book of Abstracts,
A203 P436.
Donovan WH, Dimitrijevic MR, Dahm L,
Dimitrijevic M (1982) Neurophysiological
approaches to chronic pain following spinal cord
injury. Paraplegia 20, 135
146.
Dotson RM (1997) Clinical neurophysiology laboratory tests to assess the nociceptive axis in humans.
J Clin Neurophysiol 14, 32
45.
Dowman R (2002) Pain-evoked anterior cingulate
activity generating the negative difference potential may reflect response selection processes.
Psychophysiology 39, 369
379.
Drake CG, McKenzie KG (1953) Mesencephalic
tractotomy for pain. Experience with six cases.
J Neurosurg 10, 457
462.
Drevets WC, Burton H, Videen TO et al. (1995)
Blood flow changes in human somatosensory
cortex during anticipated stimulation. Nature
373, 249
252.
Drewes AM, Andreasen A, Poulsen LH (1994)
Valproate for treatment of chronic central pain
after spinal cord injury. A double-blind cross-over
study. Paraplegia 32, 565
569.
Driver J, Mattingley JB (1998) Parietal neglect and
visual awareness. Nature Neurosci 1, 17
22.
Drouot X, Nguyen J-P, Peschanski M, Lefaucheur
J-P (2002) The antalgic efficacy of chronic
motor cortex stimulation is related to sensory

347
changes in the painful area. Brain 125,
1660
1664.
Druckman R, Lende R (1965) Central pain of spinal
cord origin. Pathogenesis and surgical relief in one
patient. Neurology 15, 518
522.
Druckman R (1966) Personal communication to
White and Sweet (1969).
Drury WL, Clarke LC (1970) Ketamine failure in
acute brain injury: a case report. Anaesth Analg
49, 859
861.
Ducreux D, Attal N, Parker F, Bouhassira D (2006)
Mechanisms of central neuropathic pain: a
combined psychophysical and fMRI study in
syringomyelia. Brain 129, 963
976.
Duncan GH, Bushnell CM, Marchand S (1991) Deep
brain stimulation: a review of basic research and
clinical studies. Pain 45, 49
59.
Duncan GH, Kupers RC, Marchand S et al. (1998)
Stimulation of human thalamus for pain
relief: possible modulatory circuits revealed by
positron emission tomography. J Neurophysiol 80,
3326
3330.
Durward QJ, Rice GP, Ball MJ, Gilbert JJ,
Kaufmann JCE (1982) Selective spinal cordectomy: clinicopathological correlation. J Neurosurg
56, 359
367.
Dworkin GF, Staats WE (1985) Posttraumatic
syringomyelia. Arch Phys Medic Rehab 66,
329
331.
Edgar RE, Best LG, Quail PA, Obert AD (1993)
Computer-assisted DREZ microcoagulation: posttraumatic spinal deafferentation pain. J Spinal Dis
6, 48
56.
Edinger L (1891) Giebt es central entstehende
Schmerzen? Dtsch Z Nervenheilk 1, 262
282.
Edmondson EA, Simpson RK, Stubler DK, Beric A
(1993) Systemic lidocaine therapy for poststroke
pain. South Med J 86, 1093
1096.
Edwards CL, Sudhakar S, Scales MT et al. (2000)
Electromyographic (EMG) biofeedback in the
comprehensive treatment of central pain and
ataxic tremor following thalamic stroke. Appl
Psychophysiol Biofeedback 25, 229
240.
Edwards WT, Habib F, Burney RG, Begin G (1985)
Intravenous lidoacine in the management of
various chronic pain states. Reg Anaesth 10, 1
6.
Eide PK, Stubhaug A, Stenehjem AE (1995) Central
dysesthesia pain after traumatic spinal cord injury
is dependent on N-methyl-D-aspartate receptor
activation. Neurosurgery 37, 1080
1087.
Eide PK, Jorum E, Stenehjem AE (1996) Somatosensory findings in patients with spinal cord injury
and central dysesthesia pain. J Neurol Neurosurg
Psych 60, 411
415.

348
Eisenach JC, Rauck RL, Curry R (2003) Intrathecal,
but not intravenous adenosine reduces allodynia
in patients with neuropathic pain. Pain 105,
65
70.
Eisenberg E, Brecker C (2002) Lumbar spinal cord
stimulation for cervical-originated central pain: a
case report. Pain 100, 299
301.
Eisenberg E, McNicol ED, Carr DB (2005) Efficacy
and safety of opioid agonists in the treatment of
neuropathic pain of nonmalignant origin. Systematic review and meta-analysis of randomized
controlled trials. JAMA 293, 3043
3052.
Ekbom KA (1966) Tegretol, a new therapy of tabetic
lightning pains. Acta Med Scand 179, 251
252.
Ekbom KA, Westerberg CE, Osterman P (1968)
Focal sensory motor seizures of spinal origin.
Lancet 1, 67
Elliott AM, Smith BH, Chambers WA (2003)
Measuring the severity of chronic pain: a research
perspective. Exp Rev Neurotherapeut 3, 581
590.
Emmers R (1981) Pain: A Spike-Interval Coded
Message in the Brain. New York: Raven Press.
Enarson MC, Hays H, Woodroffe MA (1999)
Clinical experience with oral ketamine. J Pain
Symptom Manage 17, 384
386.
Epstein FJ, Farmer JP, Freed D (1993) Adult
intramedullary spinal cord ependymomas. The
result of surgery in 38 patients. J Neurosurg 79,
204
209.
Ergenzinger ER, Glasier MM, Hahm JO, Pons TP
(1998) Cortically induced thalamic plasticity in the
primate somatosensory system. Nature Neurosci
1, 226
229.
Erickson TC, Bleckwenn WJ, Woolsey CN (1952)
Observations on the post central gyrus in relation
to pain. Trans Am Neurol Assoc 77, 57
59.
Eriksson MBE, Sjolund BH, Nielzen S (1979) Long
term results of peripheral conditioning stimulation as an analgesic measure in chronic pain.
Pain 6, 335
347.
Eriksson MBE, Sjolund BH, Sundbarg G (1984) Pain
relief from peripheral conditioning stimulation in
patients with chronic facial pain. J Neurosurg 61,
149
155.
Ervin FR, Brown CE, Mark VH (1966) Striatal
influence on facial pain. Confinia Neurologica 27,
75
86.
Espir ML, Millac P (1970) Treatment of paroxysmal
disorders in multiple sclerosis with carbamazepine
(Tegretol). J Neurol Neurosurg Psych 33, 528
531.
Eysel UT (2003) Illusions and perceived images in
the primate brain. Science 302, 789
790.
Fabritius H (1907) Studien uber die sensible Leitung
in menschlichen Ruckenmark auf grund klinischer

References
und pathologisch-anatomischer Tatsachen. Berlin:
Karger.
Falci S, Best L, Bayles R, Lammertse D, Starnes C
(2002) Dorsal root entry zone microcoagulation for spinal cord injury-related central pain:
operative intramedullary electrophysiological
guidance and clinical outcome. J Neurosurg 97,
193
200.
Falconer MA (1949) Intramedullary trigeminal
tractotomy and its place in the treatment of
facial pain. J Neurol Neurosurg Psych 12, 297
311.
Falconer MA (1953) Surgical treatment of intractable
phantom-limb pain. Br Med J 1, 299
304.
Farmer S (2002) Neural rhythms in Parkinsons
disease. Brain 125, 1175
1176.
Favre J, Taha JM, Burchiel KJ (2002) An analysis
of the respective risks of hematoma formation in
361 consecutive morphological and functional
stereotactic procedures. Neurosurgery 50, 48
56.
Fenollosa P, Pallares J, Cervera J et al. (1993)
Chronic pain in the spinal cord injured: statistical
approach and pharmacological treatment. Paraplegia 31, 722
729.
Fields HL, Adams JE (1974) Pain after cortical injury
relieved by electrical stimulation of the internal
capsule. Brain 97, 169
178.
Fields HL (1999) Pain: an unpleasant topic. Pain
(Suppl. 6), S61
S69.
Fine W (1967) Post-hemiplegic epilepsy in the
elderly. Br Med J 1, 199
201.
Finnerty GT, Connors BW (2000) Sensory deprivation without competition yields modest alterations of short-term synaptic dynamics. Proc Natl
Acad Sci USA 97, 12864
12868.
Finnerup NB, Yezierski RP, Sang CN, Burchiel KJ,
Jensen TS (2001) Treatment of spinal cord injury
pain. Pain Clin Updates IX.
Finnerup NB, Sindrup SH, Bach FW, Johannesen
IL, Jensen TS (2002) Lamotrigine in spinal cord
injury pain: a randomized controlled trial. Pain
96, 375
383.
Finnerup NB, Gottrup H, Jensen TS (2002) Anticonvulsants in central pain. Expert Opin Pharmacother 3, 1411
1420.
Finnerup NB, Johannesen IL, Bach FW, Jensen TS
(2003a) Sensory function above lesion level in
spinal cord injury patients with and without pain.
Somatosens Mot Res 20, 71
76.
Finnerup NB, Johannesen IL, Fuglsang-Frederiksen
A, Bach FW, Jensen TS (2003b) Sensory function
in spinal cord injury patients with and without
central pain. Brain 126, 57
70.
Finnerup NB, Gyldensted C, Nielsen E et al. (2003c)
MRI in chronic spinal cord injury patients

References
with and without central pain. Neurology 61,
1569
1575.
Finnerup NB, Gyldensted C, Fuglsang-Frederiksen A,
Bach FW, Jensen TS (2004) Sensory perception in
complete spinal cord injury. Acta Neurol Scand
109, 194
199.
Finnerup NB, Biering-Sorensen F, Johannesen IL
et al. (2005) Intravenous lidocaine relieves spinal
cord injury pain. A randomized controlled trial.
Anesthesiology 102, 1023
1030.
Fisher K, Hagen NA (1999) Analgesic effect of oral
ketamine in chronic neuropathic pain of spinal
origin: a case report. J Pain Symptom Manage 18,
61
66.
Fitzek S, Baumgartner U, Fitzek C et al. (2001)
Mechanisms and predictors of chronic facial pain
in lateral medullary infarction. Ann Neurol 49,
493
500.
Flor H (2003) Cortical reorganisation and chronic
pain: implications for rehabilitation. J Rehab Med
41(Suppl.), 66
72.
Foix C, Thevenard A, Nicolesco (1922) Algie
faciale dorigine bulbo-trigeminale au cours de
la Syringomyelie-Troubles Sympathiques
concomitants-Douleur a type cellulaire. Rev
Neurol 29, 990
999.
Foltz EL, White LE Jr. (1966) Rostral cingulotomy
and pain relief. In Knighton RS, Dumke PR,
eds., Pain. Henry Ford Hospital International
Symposium. Boston, MA: Little Brown,
pp. 469
491.
Forster O (1927) Die Leitungsbahnen des Schmerzgefuhls und die chirurgische Behandlung der Schmerzzustande. Berlin: Urban and Schwarzenberg.
Forster O (1936) Symptomatologie der Erkrankungen des Ruckenmarks und seiner Wurzeln. In
Bumke O, Forster O, eds., Handbuch der Neurologie, vol. 5. Berlin: Springer, pp. 1
403.
Fouad K, Pearson K (2004) Restoring walking after
spinal cord injury. Progr Neurobiol 73, 107
126.
Frank F, Tognetti F, Gaist G et al. (1982) Stereotaxic rostral mesencephalotomy in treatment of
malignant faciothoracobrachial pain syndromes.
J Neurosurg 56, 807
811.
Frank F, Frank G, Gaist G, Fabrizi A, Sturiale C
(1984) Deep brain stimulation in chronic pain
syndromes. Acta Neurochir (Wien) Suppl 33,
491
495.
Franzini A, Ferroli P, Dones I, Marras C, Broggi G
(2003) Chronic motor cortex stimulation for
movement disorders: a promising perspective.
Neurol Res 25, 123
126.
Frazier CH, Lewy FH, Rowe SN (1937) The origin
and mechanism of paroxysmal neuralgic pain and

349
the surgical treatment of central pain. Brain 610,
44
51.
Freeman LW, Heimburger RF (1947) Surgical relief
of pain in paraplegic patients. Arch Surg 55,
433
440.
Freeman W, Watts JW (1950) Psychosurgery in the
Treatment of Mental Disorders and Intractable
Pain, 2nd edn. Springfield, IL: Thomas.
Fregni F, Boggio PS, Lima MC et al. (2006) A shamcontrolled, phase II trial of transcranial direct
current stimulation for the treatment of central
pain in traumatic spinal cord injury. Pain 122,
197
209.
Friedman AH, Nashold BS (1986) DREZ lesions for
relief of pain related to spinal cord injury.
J Neurosurg 65, 465
469.
Friedman AH, Bullitt E (1988) Dorsal root
entry zone lesions in the treatment of pain
following brachial plexus avulsion, spinal cord
injury and herpes zoster. Appl Neurophysiol
51, 164
169.
Friedman AH (1991) Treatment of deafferentation
pains following peripheral nerve injuries. In
Nashold BS Jr, Ovelmen-Levitt J, eds., Deafferentation Pain Syndromes. Pathophysiology and
Treatment. New York: Raven Press.
Friehs GM, Schroettner O, Pendl G (1995) Evidence
for segregated pain and temperature conduction
within the spinothalamic tract. J Neurosurg 83,
8
12.
Frighetto L, De Salles A, Wallace R et al. (2004)
Linear accelerator thalamotomy. Surg Neurol 62,
106
113.
Frot M, Mauguiere F (2003) Dual representation of
pain in the operculo-insular cortex in humans.
Brain 126, 438
450.
Fujii M, Ohmoto Y, Kitahara T et al. (1997) Motor
cortex stimulation therapy in patients with
thalamic pain. No Shinkei Geka 25, 315
319.
Fujita T, Kitani Y (1992) Pituitary adenolysis by
electrocoagulation. In The Pain Clinic IV. VSP,
pp. 25
31.
Fujiwara N, Imai M, Nagamine T et al. (2002)
Second somatosensory area (SII) plays a significant role in selective somatosensory attention.
Cogn Brain Res 14, 389
397.
Fukaya C, Katayama Y, Yamamoto T et al. (2003)
Motor cortex stimulation in patients with poststroke pain: conscious somatosensory response
and pain control. Neurol Res 25, 153
156.
Fukuhara T, McKhann GM II, Santiago P et al.
(1999) Resolution of central pain after embolization of an arteriovenous malformation. Case
report. J Neurosurg 90, 575
579.

350
Fukui S, Shigemori S, Nosaka S (2002a) Central
pain associated with low thalamic blood flow
treated by electroconvulsive therapy. J Anaesth 16,
255
257.
Fukui S, Shigemori S, Nosaka S (2002b) A case of
central post-stroke pain with beneficial response
to electroconvulsive therapy: a proton magnetic
resonance spectroscopy study. Pain Clinic 14,
173
178.
Fukumoto M, Ushida T, Zinchuk VS, Yamamoto H,
Yoshida S (1999) Contralateral thalamic perfusion
in patients with reflex sympathetic dystrophy
syndrome. Lancet 354, 1790
1791.
Gaches J (1952) Resultat therapeutique des interventions corticales prefrontales limitees dans les
traitment de certains syndromes douloureux dit
irreductibles. Semaine des Hopitaux de Paris 28,
3667
3675.
Galaretta M, Hestrin S (1998) Frequency-dependent
synaptic depression and the balance of excitation
and inhibition in the neocortex. Nature Neurosci
1, 587
594.
Gale K (1992) GABA and epilepsy: basic concepts
from preclinical research. Epilepsia 33(Suppl. 5),
S3
S12.
Galer BS, Miller KV, Rowbotham MC (1993)
Response to intravenous lidocaine infusion differs
based on clinical diagnosis and site of nervous
system injury. Neurology 43, 1233
1235.
Gamble GE, Barberan E, Bowsher D, Tyrrell PJ,
Jones AK (2000) Post stroke shoulder pain: more
common than previously realized. Eur J Pain 4,
313
315.
Ganz E, Mullan S (1977) Percutaneous cordotomy.
In Lipton S, ed., Persistent Pain, vol. 1. London:
Academic Press, p. 31.
Garcia-Larrea L, Sindou M, Mauguiere F (1989)
Nociceptive flexion reflexes during analgesic
neurostimulation in man. Pain 39, 145
156.
Garcia-Larrea L, Peyron R, Mertens P et al. (1997)
Positron emission tomography during motor
cortex stimulation for pain control. Stereotact
Funct Neurosurg 68, 141
148.
Garcia-Larrea L, Peyron R, Mertens P et al. (1999)
Electrical stimulation of motor cortex for pain
control: a combined PET-scan and electrophysiological study. Pain 83, 259
273.
Garcia-Larrea L, Convers P, Magnin M et al. (2000)
Laser-evoked potential abnormalities in central
pain patients: the influence of spontaneous and
provoked pain. Brain 125, 2766
2781.
Garcin R (1937) La douleur dans les affections
organiques du syste`me nerveux central. Rev Neurol
68, 105
153.

References
Garcin R, Lepresle J (1954) Syndrome sensitif de
type thalamique et a topographie cheiro-orale par
lesion localisee du thalamus. Rev Neurol 90,
124
129.
Garcin R (1957) La douleur dans les affections
du syste`me nerveux central (thalamus, region
bulbo-protuberantielle). In Alajouanine Th, ed.,
La douleur et les douleurs. Paris: Masson,
pp. 199
213.
Garcin R (1968) Thalamic syndrome and pain
of central origin. In Soulairac A, Cahn J,
Charpentier J, eds., Pain. London: Academic
Press, pp. 521
539.
Gardner WJ, Karnosh LJ, McLure CC (1955)
Residual function following hemispherectomy
for tumour and for infantile hemiplegia. Brain
78, 487
502.
Gatscher S, Becker R, Uhle E, Bertalanffy H (2001)
Combined intrathecal baclofen and morphine
infusion for the treatment of spasticity related
pain and central deafferentation pain. Acta
Neurochir 79(Suppl.), 75
76.
Gazzaniga MS (1998) The split brain revisited.
Sci Am 279, 50
55.
Getting PA (1989) Emerging principles governing
the operation of neural networks. Annu Rev
Neurosci 12, 185
204.
Gharabaghi A, Hellwig D, Rosahl SK et al. (2005)
Volumetric image guidance for motor cortex
stimulation: integration of three-dimensional
cortical anatomy and functional imaging. Neurosurgery 57(ONS Suppl. 1), 114
120.
Gibson JC, White LC (1971) Denervation hyperpathia: a convulsive syndrome of the spinal cord
responsive to carbamazepine therapy. J Neurosurg
35, 287
290.
Gimenez-Roldan S, Martin M (1981) Tabetic lightning pains: high-dosage intravenous penicillin
versus carbamazepine therapy. Eur Neurol 20,
424
428.
Gioia DF, Wallace PB, Fuste FJ, Greene M (1967)
A stereotaxic method of surgery for the relief of
intractable pain. Int Surg 48, 409
416.
Glees P, Bailey EA (1951) Schichtung und
Fasergroesse des tractus spinothalamicus des
Menschen. Mschr Psychiatr Neurol 122,
129
141.
Glynn CJ, Jamous MA, Teddy PJ, Moore RA,
Lloyd JW (1986) Role of spinal noradrenergic
system in transmission of pain in patients with
spinal cord injury. Lancet ii, 1249
1250.
Glynn CJ, Jamous MA, Teddy PJ (1992) Cerebrospinal fluid kinetics of epidural clonidine in man.
Pain 49, 361
367.

References
Gold L, Lauritzen M (2002) Neuronal deactivation
explains decreased cerebellar blood flow in
response to focal cerebral ischemia or suppressed
neocortical function. Proc Natl Acad Sci USA 99,
7699
7704.
Goldman-Rakic PS (1995) Cellular basis of working
memory. Neuron 14, 477
485.
Gonzales GR, Herskovitz S, Rosenblum M et al.
(1992) Central pain from cerebral abscess: thalamic syndrome in AIDS patients with toxoplasmosis. Neurology 42, 1107
1109.
Gonzales GR, Lewis SA, Weaver AL (2001) Tactile
illusion perception in patients with central pain.
Mayo Clin Proc 76, 267
274.
Gonzales GR, Tuttle SL, Thaler HT, Manfredi PL
(2003) Central pain in cancer patients. J Pain 4,
351
354.
Gorecki J, Hirayama T, Dostrovsky JO, Tasker RR,
Lenz FA (1989) Thalamic stimulation and recording in patients with deafferentation and central
pain. Stereotact Funct Neurosurg 52, 219
226.
Gorecki JP, Nashold BS (1995) The Duke experience
with the nucleus caudalis DREZ operation.
Acta Neurochir (Wien) Suppl 64, 128
131.
Graf C (1960) Consideration in loss of sensory level
after bilateral cervical cordotomy. Arch Neurol 3,
410
415.
Graff-Radford NR, Damasio H, Yamada T, Eslinger
PJ, Damasio AR (1985) Nonhaemorrhagic thalamic infarction. Clinical, neuropsychological and
electrophysiological findings in four anatomical
groups defined by computerized tomography.
Brain 108, 485
516.
Grant FC (1948) Complications accompanying
surgical relief of pain in trigeminal neuralgia.
Am J Surg 75, 42
47.
Grant FC, Weinberger LH (1941) Experiences with
intramedullary tractotomy: IV. Surgery of the
brain-stem and its operative complications. Surg
Gyn Obst 72, 747
754.
Grant FC, Wood FA (1958) Experiences with
cordotomy. Clin Neurosurg 5, 38
65.
Greenspan JD, Joy SE, McGillis SLB, Checkosky
CM, Bolanowski SJ (1997) A longitudinal study
of somesthetic perceptual disorders in an individual with a unilateral thalamic lesion. Pain 72,
13
25.
Greenspan JD, Ohara S, Sarlani E, Lenz FA (2004)
Allodynia in patients with post-stroke central
pain (CPSP) studied by statistical quantitative
sensory testing within individuals. Pain 109,
357
366.
Greiff F (1883) Zur localisation der Hemichorea.
Arch Psychol Nervenkrankheit 14, 598

351
Guecer G, Niedermeyer E, Long MD (1978)
Thalamic recordings in patients with chronic
pain. J Neurol 219, 47
61.
Guillaume J, De Seze S, Mazars G (1949) Chirurgie
cerebro-spinale de la douleur. Paris: Presses Univ
France.
Guillery RW, Feig SL, Lozsadi DA (1998) Paying
attention to the thalamic reticular nucleus. TINS
21, 28
32.
Gupta A, Wang Y, Markram H (2000) Organizing
principles for a diversity of GABAergic interneurons and synapses in the neocortex. Science 287,
273
278.
Gybels JM, Sweet WH (1989) Neurosurgical treatment of persistent pain. In Gildenberg PL, ed.,
Pain and Headache. Basel: Karger, pp. 1
253.
Gybels J, Kupers R, Nuttin B (1993) Therapeutic
stereotactic procedures on the thalamus for pain.
Acta Neurochir 124, 19
22.
Gybels JM, Kupers RC (1995) Brain stimulation in
the management of persistent pain. In Schmideck
HH, Sweet WH, eds., Operative Neurosurgical
Techniques. Indications, Methods and Results,
3rd edn. Philadelphia, PA: WB Saunders,
pp. 1389
1398.
Haber SN, Gdowski MJ (2004) The basal ganglia.
In Paxinos G, Mai JK, eds., The Human Nervous
System, 2nd edn. Amsterdam: Elsevier Academic
Press, pp. 677
738.
Hachen HJ (1978) Psychological, neurophysiological
and therapeutic aspects of chronic pain: preliminary results with transcutaneous electrical stimulation. Paraplegia 15, 353
367.
Hagelberg N, Martikainen IK, Mansikka H et al.
(2002) Dopamine D2 receptor binding in the
human brain is associated with the response to
painful stimulation and pain modulatory capacity.
Pain 99, 273
279.
Haines DR, Gaines SP (1999) N-of-1 randomised
controlled trials of oral ketamine in patients with
chronic pain. Pain 83, 283
287.
Hallin RG, Wiesenfeld-Hallin Z (1983) Does
sympathetic activity modify afferent inflow at
the receptor level in man? J Auton Nerv Syst 7,
391
397.
Hamamci N, Dursun E, Ural C, Cakci A (1996)
Calcitonin treatment in reflex sympathetic dystrophy: a preliminary study. Br J Clin Pract 50,
373
375.
Hamby WB, Shinners BM, Marsh IA (1948)
Trigeminal tractotomy. Observations on fortyeight cases. Arch Surg (Lond) 57, 171
177.
Hamby WB (1961) Reversible central pain. Arch
Neurol 5, 528
532.

352
Hampf G, Bowsher D (1989) Distigmine and
amitriptyline in the treatment of chronic pain.
Anesth Progr 36, 58
62.
Hanajima R, Ashby P, Lang AE, Lozano AM (2002)
Effects of acute stimulation through contacts
placed on the motor cortex for chronic stimulation. Clin Neurophysiol 113, 635
641.
Hankinson J (1962) Neurosurgical aspects of relief of
pain at the cerebral level. In Keele CA, Smith R,
eds., The Assessment of Pain in Man and Animals.
Edinburgh: Livingston, pp. 135
143.
Hannington-Kiff LG (1974) Intravenous regional
sympathetic block with guanethidine. Lancet 1,
1019
1020.
Hansebout RR, Blight AR, Fawcett S, Reddy K
(1993) 4-Aminopyridine in chronic spinal cord
injury: a controlled, double-blind, cross-over
study in eight patients. J Neurotrauma 10, 1
18.
Hansson P (2004) Post-stroke pain case study:
clinical characteristics, therapeutic options and
long-term follow-up. Eur J Neurol 11(Suppl. 1),
22
30.
Harbison J, Dennehy F, Keating D (1997) Lamotrigine for pain with hyperalgesia. Ir Med J 90, 56
Harano K, Koga A, Takasaki M, Totoki T (1999)
Intracisternal methylprednisolone administration
for thalamic and other intractable pains. In 9th
World Congress on Pain, Book of Abstracts. Seattle,
WA: IASP Press, A234, p. 445.
Harden RN, Brenman E, Saltz S, Houle T (2002)
Topiramate in the management of spinal cord
injury pain: a double-blind, randomised, placebocontrolled pilot study. Progr Pain Res Manage
23, 393
407.
Harrington M, Bone I (1981) Spinal meningioma
presenting as focal epilepsy; a case report. Br Med J
282, 1984
1985.
Hariz MI, Bergenheim AT (1995) Thalamic stereotaxic for chronic pain: ablative lesion or stimulation? Stereotact Funct Neurosurg 64, 47
55.
Hassenbusch SJ, Stanton-Hicks M, Covington EC,
Walsh JG, Guthrey DS (1995) Long-term intraspinal infusions of opioids in the treatment of neuropathic pain. J Pain Symptom Manage 10, 527
543.
Hassler R, Riechert T (1959) Klinische und
anatomische Befunde bei sterotaktischen
Schmerz-operationen im Thalamus. Arch Psychiat
200, 93
122.
Hassler R (1960) Die zentrale Systeme des
Schmerzes. Acta Neurochir 8, 365
423.
Hassler R (1970) Dichotomy of facial pain conduction in the diencephalon. In Hassler R, Walker AE,
eds., Trigeminal Neuralgia. Philadelphia, PA:
WB Saunders, pp. 123
138.

References
Hawking S (1988) A Brief History of Time. London:
Bantam.
Hayashi M, Taira T, Ochiai T et al. (2005) Gamma
knife surgery of the pituitary: new treatment for
thalamic pain syndrome. J Neurosurg 102(Suppl.),
38
41 (includes Hayashi et al. (2003) Stereotact
Funct Neurosurg 81, 75
83).
Hayes KC, Potter PJ, Wolfe DL et al. (1994)
4-Aminopyridine-sensitive neurologic deficits in
patients with spinal cord injury. J Neurotrauma
11, 433
446.
Head H, Holmes G (1911) Researches into sensory
disturbances from cerebral lesions. Brain 34,
102
254.
Hecaen H, Talairach J, David M, Dell MB (1949)
Coagulations limitees du thalamus dans les algies
du syndrome thalamique. Resultats therapeutiques et physiopathologiques. Rev Neurol 81,
917
931.
Heilporn A (1978) Two therapeutic experiments on
stubborn pain in spinal cord lesions: coupling
melitracen-flupenthixol and the transcutaneous
nerve stimulation. Paraplegia 15, 368
372.
Heiskanen T, Hartel B, Dahl ML, Seppala T, Kalso E
(2002) Analgesic effects of dextromethorphan and
morphine in patients with chronic pain. Pain 96,
261
267.
Heiss WD, Pawlik G, Herholz K et al. (1986)
Remote metabolic disturbances in lesions of brain
stem and diencephalons: a PET study. In Samii M,
ed., Surgery In and Around the Brain Stem and
the Third Ventricle. Berlin: Springer Verlag,
pp. 207
212.
Helfant MH, Leksell L, Strang RR (1965) Experiences
with intractable pain treated by sterotaxic mesencephalotomy. Acta Chir Scand 129, 573
580.
Helmchen C, Lindig M, Petersen D, Tronnier V
(2002) Disappearance of central thalamic pain
syndrome after contralateral parietal lobe lesion:
implications for therapeutic brain stimulation.
Pain 98, 325
330.
Henderson JM, Boongird A, Rosenow JM,
LaPresto E, Rezai AR (2004) Recovery of pain
control by intensive reprogramming after loss of
benefit from motor cortex stimulation for neuropathic pain. Stereotact Funct Neurosurg 82,
207
213.
Henkel K, Bengel D (2005) Gekreuztes zentralneuropatisches Schmerzsyndrom nach bakterieller
Meningoenzephalitis. Der Schmerz 19, 55
58.
Herman RM, DLuzansky SC, Ippolito R (1992)
Intrathecal baclofen suppresses central pain in
patients with spinal lesions. A pilot study. Clin J.
Pain 8, 338
345.

References
Herregodts P, Stadnik T, Deridder F, DHaens J
(1995) Cortical stimulation for central neuropathic pain: 3-D surface MIZI for easy determination of the motor cortex. Acta Neurochir (Wien)
Suppl 64, 132
135.
Hida K, Iwasaki Y, Imamura H, Abe H (1994)
Postraumatic syringomyelia: its characteristics,
magnetic resonance imaging findings and surgical
management. Neurosurgery 36, 886
891.
Hirato M, Kawashima Y, Shibasaki T, Ohye C (1991)
Pathophysiology of central (thalamic) pain: a possible role of the intralaminar nuclei in superficial
pain. Acta Neurochir (Wien) Suppl 52, 133
136.
Hirato M, Horikoshi S, Kawashima Y et al. (1993)
The possible role of the cerebral cortex adjacent to
the central sulcus for the genesis of central
(thalamic) pain: a metabolic study. Acta Neurochir
(Wien) Suppl 58, 141
144.
Hirato M, Ohye C, Shibazaki T et al. (1995) Gamma
knife thalamotomy for the treatment of functional
disorders. Stereotact Funct Neurosurg 64(Suppl. 1),
164
171.
Hirayama T, Dostrovsky JO, Gorecki J, Tasker RR,
Lenz FA (1989) Recordings of abnormal activity
in patients with deafferentation and central pain.
Stereotact Funct Neurosurg 52, 120
126.
Hitchcock ER, Teixeira MJ (1981) A comparison of
results from center-median and basal thalamotomies for pain. Surg Neurol 15, 341
351.
Hodge CJ Jr., King RB (1976) Medical modification
of sensation. J Neurosurg 44, 21
28.
Hofbauer RK, Fiset P, Plourde G, Backman SB,
Bushnell MC (2004) Dose-dependent effects of
propofol on the central processing of thermal
pain. Anesthesiology 100, 386
394.
Hoffmann W (1933) Thalamussyndrom auf grund
einer kleinen Lesion. Psychol Neurol 54, 362
374.
Holden C (2004) Imaging studies show how brain
thinks about pain. Science 303, 1121.
Holland CT, Charles JA, Smith SH, Cortaville PE
(2000) Hemihyperaesthesia and hyperresponsiveness resembling central pain syndrome in a dog
with a forebrain oligodendroglioma. Aust Vet J
78, 676
680.
Holmes G (1919) Pain of central origin. In Osler W,
ed., Contributions to Medical and Biological
Research. New York: Paul B Hoeber, pp. 235
246.
Holmgren J, Leijon G, Boivie J, Johansson I, Ilievska
L (1990) Central post-stroke pain: somatosensory
evoked potentials in relation to location of the
lesion and sensory signs. Pain 40, 43
52.
Honey CR, Stoessl J, Tsui JK, Schulzer M, Calne DB
(1999) Unilateral pallidotomy for reduction of
parkinsonian pain. J Neurosurg 91, 198
201.

353
Hoogenraad TU, Ramos LMP, Van Gijn J (1994)
Visually induced central pain and arm withdrawal
after right parietal lobe infarction. J Neurol
Neurosurg Psych 57, 850
852.
Horrax G (1946) Experiences with cortical excisions
for the relief of intractable pain in the extremities.
Surgery 20, 593
602.
Horrax G, Lang E (1957) Complications of chordotomy. Surg Clin N Am 37, 849
854.
Hosobuchi Y (1986) Subcortical electrical stimulation for control of intractable pain in humans.
Report of 122 cases (1970
1984). J Neurosurg 64,
543
553.
Hosobuchi Y (1990) Alpha-methyldopa blocks the
analgesic effect of sensory thalamic stimulation in
humans. Pain 5(Suppl.), S274.
Hosobuchi Y (1993) Motor cortical stimulation
for control of central deafferentation pain. In
Devinsky O, Beric A, Dogari M, eds., Electrical
and Magnetic Stimulation of the Brain and Spinal
Cord, Advances in Neurology Series. New York:
Raven Press, pp. 215
217.
Houtchens MK, Richert JR, Sami A, Rose JW (1997)
Open label gabapentin treatment for pain in
multiple sclerosis. Mult Scler 3, 250
253.
Hsieh J-C, Belfrage M, Stone-Elander S, Hansson P,
Ingvar M (1995) Central representation of
chronic ongoing neuropathic pain studies by
positron emission tomography. Pain 63,
225
236.
Hua SE, Garonzik IM, Lee JI, Lenz FA (2000)
Microelectrode studies of normal organization
and plasticity of human somatosensory thalamus.
J Clin Neurophysiol 17, 559
574.
Hughes JT (1976) Diseases of the spine and
spinal cord. In Blackwood W, Corsellis J, eds.,
Greenfields Neuropathology. London: Arnold,
pp. 652
687.
Huguenard JR, Prince DA (1997) Basic mechanisms
of epileptic discharges in the thalamus. In Steriade
M, Jones EG, McCormick DA, eds., Thalamus: II.
Experimental and Clinical Aspects. Amsterdam:
Elsevier, pp. 295
330.
Hunt WE, Goodman JH, Bingham WG (1975)
Stimulation of the dorsal spinal cord for treatment
of intractable pain: a preliminary report. Surg
Neurol 4, 153
156.
Hutcheon B, Yarom Y (2000) Resonance, oscillation
and the intrinsic preference of neurons. TINS 23,
216
222.
Hutchison WD, Davis KD, Lozano AM, Tasker RR,
Dostrovsky JO (1999) Pain-related neurons in
the human cingulate cortex. Nature Neurosci 2,
403
405.

354
Hyndman OR (1942) Lissauers tract section. A
contribution to chordotomy for the relief of pain.
Preliminary report. J Int Coll Surg 5, 394
400.
Iacono RP, Nashold BS Jr. (1982) Mental and
behavioral effects of brain stem and hypothalamic
stimulation in man. Human Neurobiol 1, 273
279.
Iannetti GD, Truini A, Romaniello A et al. (2003)
Evidence of a specific spinal pathway for the
sense of warmth in humans. J Neurophysiol 89,
562
570.
Iwamura Y (1998) Hierarchical somatosensory
processing. Curr Opin Neurobiol 8, 522
528.
Iwamura Y, Iriki A, Tanaka M (1994) Bilateral hand
representation in the postcentral somatosensory
cortex. Nature 369, 554
556.
Izhikevich EM, Desai NS, Walcott EC, Hoppensteadt
FC (2003) Bursts as a unit of neural information:
selective communication via resonance. TINS 26,
161
167.
Jacobs KM, Donoghue JP (1991) Reshaping the
cortical motor map by unmasking latent intracortical connections. Science 251, 944
947.
Janig W, Koltzenburg M (1991) What is the
interaction between the sympathetic terminal
and the primary afferent fiber? In Basbaum AI,
Besson JM, eds., Towards a New Pharmacotherapy
of Pain. Chichester: John Wiley, pp. 331
335.
Janis KM, Wright W (1972) Failure to produce
analgesia with ketamine in two patients with
cortical disease. Anesthesiology 36, 405
406.
Jasmin L, Tien D, Janni G, Ohara PT (2003) Is
noradrenaline a significant factor in the analgesic
effect of antidepressants? Pain 106, 3
8.
Jeanmonod D, Magnin M, Morel A (1996) Lowthreshold calcium spike bursts in the human
thalamus. Common physiopathology for sensory,
motor and limbic positive symptoms. Brain 119,
363
375.
Jeanmonod D, Magnin M, Morel A, Siegemund M
(2001) Surgical control of the human thalamocortical dysrhythmia: I Central lateral thalamotomy in neurogenic pain. Thalam Rel Syst 1,
71
79.
Jefferys JGR, Traub RD, Whittington MA (1996)
Neuronal networks for induced 40Hz rhythms.
TINS 19, 202
208.
Jefferson A (1983) Cordectomy for intractable pain
in paraplegia. In Lipton S, Miles J, eds., Persistent
Pain: Modern Methods for Treatment, vol. 4.
London: Grune and Stratton, pp. 115
132.
Jefferson A (1987) Personal communication to
Gybels and Sweet (1989).
Jensen TS, Gottrup H, Johansen P et al. (1999)
Positron emission tomography (PET) and

References
psychophysical studies in post-stroke pain. In
9th World Congress on Pain, Book of Abstracts.
Seattle, WA: IASP Press, A77.
Jensen TS, Gottrup H, Kristensen AD, Andersen G,
Vestergaard K (2002) Cold sensation in central
post-stroke pain. In 10th World Congress on Pain,
Book of Abstracts. Seattle, WA: IASP Press,
A95-P91.
Johnson RE, Kanigsberg ND, Jimenez CL (2000)
Localized pruritus: a presenting symptom of a
spinal cord tumor in a child with features of
neurofibromatosis. J Am Acad Dermatol 43,
958
961.
Jones AK, Qi LY, Fujirawa T et al. (1991) In vivo
distribution of opioid receptors in man in relation
to the cortical projections of the medial and lateral
pain systems measured with positron emission
tomography, Neurosci Lett 126, 25
28.
Jones AKP, Watabe H, Cunningham VJ, Jones T
(2004) Cerebral decreases in opioid receptor
binding in patients with central neuropathic
pain measured by [11C] diprenorphine binding
and PET. Eur J Pain 8, 479
485.
Jones EG (2001) The thalamic matrix and thalamocortical synchrony. TINS 24, 595
601.
Joyner J, Mealy J Jr., Freeman LW (1966) Cordotomy for intractable pain of non malignant origin.
Arch Surg 93, 480
486.
Juliano SL, Ma W, Eslin D (1991) Cholinergic
depletion prevents expansion of topographic maps
in somatosensory cortex. Proc Natl Acad Sci USA
88, 780
784.
Kaas JH (1999) Is most of neural plasticity in the
thalamus cortical? Proc Natl Acad Sci USA 96,
7622
7623.
Kaas JH (2004) Somatosensory system. In Paxinos G,
Mai JK, eds., The Human Nervous System,
2nd edn. Amsterdam: Elsevier Academic Press,
pp. 1061
1093.
Kabirov EI, Staroselseva NG (2002) Transcutaneous
electrostimulation in the treatment of dysesthetic
pain in syringomyelia. In 10th World Congress on
Pain, Book of Abstracts. Seattle, WA: IASP Press,
A1656-P204.
Kakulas BA, Smith E, Gaekwad U, Kaeln C, Jacobsen
PF (1990) The neuropathology of pain and
abnormal sensation in human spinal cord injury
derived from the clinicopathological data base of
the Royal Perth Hospital. In Dimitrijevic MR,
Wall PD, Lindblom U, eds., Recent Achievements
in Restorative Neurology: 3. Altered Sensation and
Pain. Basel: Karger, p. 38.
Kalman S, Osterberg A, Sorensen J, Boivie J,
Bertler A (2002) Morphine responsiveness in

References
a group of well-defined multiple sclerosis patients:
a study with i.v. morphine. Eur J Pain 6, 69
80.
Kameda W, Kawanami T, Kurita K et al. for the
Study Group of the Association of Cerebrovascular Disease in Tohoku (2004) Lateral and medial
medullary infarction. A comparative analysis of
214 patients. Stroke 35, 694
699.
Kameyama M (1976
77) Vascular lesions of the
thalamus on the dominant and nondominant side.
Appl Neurophysiol 39, 171
177.
Kapadia NP, Harden N (2000) Gabapentin for
chronic pain in spinal cord injury: a case report.
Arch Phys Med Rehabil 81, 1439
1441.
Katayama Y, Fukaya C, Yamamoto T (1998)
Poststroke pain control by chronic motor cortex
stimulation: neurological characteristics predicting a favorable response. J Neurosurg 89, 585
591.
Katayama Y, Yamamoto T, Kobayashi K et al. (2001)
Motor cortex stimulation for post-stroke pain:
comparison of spinal cord and thalamic stimulation. Stereotact Funct Neurosurg 77, 183
186.
Kawahara N, Sato K, Muraki M et al. (1986) CT
classification of small thalamic hemorrhages and
their clinical implications. Neurology 36, 165
172.
Kendall D (1939) Some observations on central pain.
Brain 62, 253
273.
Khedr EM, Kotb H, Kamel NF et al. (2005) Longlasting antalgic effects of daily sessions of repetitive transcranial magnetic stimulation in central
and peripheral neuropathic pain. J Neurol Neurosurg Psych 76, 833
838.
Kim D, Park D, Choi S et al. (2003) Thalamic
control of visceral nociception mediated by T-type
Ca2 channels. Science 302, 117
119.
Kim H, Neubert JK, San Miguel A et al. (2004)
Genetic influence on variability in human acute
experimental pain sensitivity associated with
gender, ethnicity and psychological temperament.
Pain 109, 488
496.
Kim JS, Lee JH, Lee MC (1995) Sensory changes
in the ipsilateral extremity. A clinical variant of
lateral medullary infarction. Stroke 26, 1956
1958.
Kim JS, Kim HG, Chung CS (1995) Medial
medullary syndrome. Report of 18 new patients
and a review of the literature. Stroke 26,
1548
1552.
Kim JS, Lee JH, Im JH, Lee MC (1995) Syndromes
of pontine base infarction. A clinical-radiological
correlation study. Stroke 26, 950
955.
Kim JS (1996a) Bilateral perioral sensory symptom
after unilateral stroke: does it have a localizing
value? J Neurol Sci 140, 123
128.
Kim JS (1996b) Restricted nonacral sensory syndrome. Stroke 27, 988
990.

355
Kim JS (1998) Delayed-onset ipsilateral sensory symptoms in patients with central post-stroke pain. Eur
Neurol 40, 201
206.
Kim JS (1999) Aggravation of poststroke sensory
symptoms after a second stroke on the opposite
side. Eur Neurol 42, 200
204.
Kim JS (2003) Central post-stroke pain or paresthesia in lenticulocapsular hemorrhages. Neurology 61, 679
682.
Kim JS, Bae YH (1997) Pure or predominant sensory
stroke due to brain stem lesion. Stroke 28,
1761
1764.
Kim JS, Choi-Kwon S (1996) Discriminative sensory
dysfunction after unilateral stroke. Stroke 27,
677
682.
Kim JS, Choi-Kwon S (1999) Sensory sequelae
of medullary infarction: differences between lateral
and medial medullary syndrome. Stroke 30,
2697
2703.
Kim SH, Tasker RR, Oh MY (2001) Spinal cord
stimulation for nonspecific limb pain versus
neuropathic pain and spontaneous versus evoked
pain. Neurosurgery 48, 1056
1065.
Kim LJ, Klopfenstein JD, Zabramski JM, Sonntag
VK, Spetzler RF (2006) Analysis of pain resolution
after surgical resection of intramedullary spinal
cord cavernous malformations. Neurosurgery
58, 106
111.
Kimyai-Asadi A, Nousari HC, Kimyai-Asadi T,
Dilani F (1999) Poststroke pruritus. Stroke
30, 692.
King CA, Huff FJ, Jorizzo JL (1982) Unilateral
neurogenic pruritus: paroxysmal itching associated with central nervous system lesions.
Ann Intern Med 97, 222
223.
King RB (1977) Anterior commissurotomy for
intractable pain. J Neurosurg 47, 7
11.
Kingery WS (1997) A critical review of controlled
clinical trials for peripheral neuropathic pain
and complex regional pain syndromes. Pain 73,
123
139.
Kinnier Wilson SA (1927) Dysesthesias and their
neural correlates. Brain 50, 428
461.
Kirchner A, Birklein F, Stefan H, Handwerker HO
(2001) Vagus nerve stimulation: a new option for
the treatment of chronic pain syndromes. Schmerz
15, 272
277.
Kiriakopoulos ET, Tasker RR, Nicosia S, Wood ML,
Mikulis DJ (1997) Functional magnetic resonance
imaging: a potential tool for the evaluation of
spinal cord stimulation. Technical case report.
Neurosurgery 41, 501
504.
Kiss ZHT, Dostrovsky JO, Tasker RR (1994)
Plasticity in human somatosensory thalamus

356
as a result of deafferentation. Stereotact Funct
Neurosurg 62, 153
163.
Kiss ZHT, Davis KD, Tasker RR, Lozano AM,
Dostrovsky JO (1995) Human thalamic neurons
develop novel receptive fields within minutes of
deafferentation. J Neurosurg 82, 373A (A841).
Kittler JT, Moss SJ (2003) Modulation of GABA-A
receptor activity by phosphorylation and receptor
trafficking: implications for the efficacy of synaptic inhibition. Curr Opin Neurobiol 13, 341
347.
Knecht S, Henningsen H, Elbert T et al. (1996)
Reorganizational and perceptional changes after
amputation. Brain 119, 1214
1219.
Knecht S, Soros P, Guertler S et al. (1998) Phantom
sensations following acute pain. Pain 77,
209
213.
Kohlhoff H, Lorenz J, Scharein E et al. (1999)
Cortical disorganization in patients with poststroke pain studied by MEG. In 9th World
Congress on Pain, Book of Abstracts. Seattle, WA:
IASP Press, A27.
Koehling R (2002) GABA becomes exciting. Science
298, 1350
1352.
Koltzenburg M, Wall PD, McMahon SB (1999)
Does the right know what the left is doing?
TINS 22, 122
127.
Kondziolka D, Wechsler L, Achim C (2002) Neural
transplantation for stroke. J Clin Neurosci 9,
225
230.
Kong K-H, Woon V-C, Yang S-Y (2004) Prevalence
of chronic pain and its impact on health-related
quality of life in stroke survivors. Arch Phys Med
Rehabil 85, 35
40.
Koszewski W, Jarosz J, Pernak-De Gast J (2003)
Stereotactic posterior capsulo-lentiform deafferentation as an effective treatment in central
post-stroke pain. A new surgical method for
intractable central pain control? Pain Clinic 15,
115
123.
Koyama T, Arakawa Y, Shibata M, Mashimo T,
Yoshiya I (1998) Effect of barbiturate on central
pain: difference between intravenous administration and oral administration. Clin J Pain 14,
86
88.
Krainick JU, Thoden U (1989) Spinal cord stimulation. In Wall PD, Melzack R, eds., Textbook of
Pain, 2nd edn. Edinburgh: Churchill Livingstone,
pp. 920
924.
Kramer KM, Levine AM (1997) Posttraumatic
syringomyelia. Clin Orthoped Rel Res 334,
190
199.
Krauss JK, Pohle T, Weigel R, Kalbarzcyk A (2001)
Somatosensory thalamic stimulation versus
center-median-parafascicular complex stimulation

References
in 11 patients with neuropathic pain. Stereotact
Funct Neurosurg 77, 194.
Kringelbach ML, Rolls ET (2004) The functional
neuroanatomy of the human orbitofrontal cortex:
evidence from neuroimaging and neuropsychology. Progr Neurobiol 72, 341
372.
Krueger AG (1960) Management of painful states
in injuries of the spinal cord and cauda equina.
Am J Phys Med 39, 103
110.
Krupa DJ, Wiest MC, Shuler MG, Laubach M,
Nicolelis MAL (2004) Layer-specific somatosensory cortical activation during active tactile
discrimination. Science 304, 1989
1992.
Kudo T, Yoshii N, Shimizu S et al. (1968) Sterotaxic
thalamotomy for pain relief. Tohoku J Exp Med 96,
219
234.
Kuhl DE, Phelps ME, Kowell AP et al. (1980) Effects
of stroke on local cerebral metabolism and
perfusion: mapping by emission computed tomography of 18FDG and 13NH3. Ann Neurol 8,
47
60.
Kuhner A (1981) Percutaneous cordotomy. Actual
situation in pain surgery. Anesth Analg (Paris) 38,
357
359.
Kuhn WG (1947) The care and rehabilitation of
patients with injuries of the spinal cord and cauda
equina. A preliminary report on 133 cases.
J Neurosurg 4, 40
68.
Kumagai Y, Taga K, Hokari T et al. (1990) The effect
of DREZ (dorsal root entry zone) lesions on
intractable pain in patients with spinal cord
injury. Masui 39, 632
638.
Kumar K, Toth C, Nath RK (1997) Deep brain
stimulation for intractable pain: a 15-year experience. Neurosurgery 40, 736
747 (updates Can J
Surg 1985, 28, 20
22, Neurosurgery 1990, 26,
774
782).
Kumar K, Hunter G, Demeria D (2006) Spinal cord
stimulation in treatment of chronic benign pain:
challenges in treatment planning and present
status, a 22-year experience. Neurosurgery 58,
481
496 (updates J Neurosurg 1991, 75,
402
407, Surg Neurol 1998, 50, 110
121).
Kumral E, Kocaer T, Ertubey N, Kumral K (1995)
Thalamic hemorrhage. A prospective study of 100
patients. Stroke 26, 964
970.
Kumral E, Celebisoy N (1996) Visually evoked
hyperpathia due to thalamic hemorrhage: a
variant of Dejerine
Roussy syndrome. Stroke 27,
774
775.
Kumral E, Evyapan D, Balkir K, Kutluhan S (2001)
Bilateral thalamic infarction. Clinical, etiological
and MRI correlates. Acta Neurol Scand 103,
35
42.

References
Kuong PG (1984) Practical Neuroanatomy and
Neurological Syndromes. Beijing: Peoples Army
Medical Publishing.
Kupers RC, Konings H, Adriaensen H, Gybels JM
(1991) Morphine differentially affects the sensory
and affective pain ratings in a neurogenic and
idiopathic forms of pain. Pain 47, 5
12.
Kuroda R, Yamada Y, Kondo S et al. (2000) Electrical
stimulation of the second somatosensory cortex for
intractable pain: a case report and experimental
study. Stereotact Funct Neurosurg 74, 226.
Kuzniecki R, Pan J, Martin R et al. (2002)
Modulation of cerebral GABA by topirate,
lamotrigine and gabapentin in healthy humans.
Neurology 58, 368
372.
Kvarnstrom A, Karlsten R, Quiding H, Gordh T
(2004) The analgesic effect of intravenous ketamine and lidocaine on pain after spinal cord
injury. Acta Anaesthesiol Scand 48, 498
506.
Lahuerta J, Bowsher D, Lipton S, Buxton PH
(1994) Percutaneous cervical cordotomy: a review
of 181 operations on 146 patients with a study
on the location of pain fibers in the C-2 spinal
cord segment of 29 cases. J Neurosurg 80,
975
985.
Laine E, Gros C (1956) Lhemispherectomie. Reunion
annuelle, Soc Neurochir Langue Franc. Paris:
Masson et Cie.
Laitinen LV (1977) Anterior pulvinotomy in the
treatment of intractable pain. In Sweet WH,
Obrador S, Martin-Rodriguez JG, eds., Neurosurgical Treatment in Psychiatry, Pain, and
Epilepsy. Baltimore, MD: University Park Press,
pp. 669
672.
Laitinen LV (1988) Mesencephalotomy and thalamotomy for chronic pain. In Lunsford LD, ed.,
Modern Sterotactic Neurosurgery. Boston, MA:
Martinus Nijhoff, pp. 269
277.
Lampl Y, Gilad R, Eshel Y, Sarova-Pinhas I (1995)
Neurological and functional outcome in patients
with supratentorial hemorrhages. A prospective
study. Stroke 26, 2249
2253.
Lampl C, Yazdi K, Roeper C (2002) Amitriptyline
in the prophylaxis of central post-stroke
pain. Preliminary results of 39 patients in a
placebo-controlled, long-term study. Stroke 33,
3030
3032.
Lance JW (1996) The red ear syndrome. Neurology
47, 617
620.
Lapresle J, Guiot G (1953) E`tude des resultats
eloignes et en particulier des sequelles neurologiques a` type de douleur central dans 8 cas de
cordotomie antero-laterale pour coxarthrose.
Semaine des Hopitaux de Paris 29, 2189
2198.

357
Larson SJ, Sances A, Riegel DH et al. (1974)
Neurophysiological effects of dorsal column
stimulation in man and monkey. J Neurosurg 41,
217
223.
Laspiur RD (1956) Estereotaxis. En relacion a dos
casos operados de sindrome talamico vascular.
Arch Inst Neurochir 97, 113.
La Terre EC, De Volder AG, Goffinet AM (1988)
Brain glucose metabolism in thalamic syndrome.
J Neurol Neurosurg Psych 51, 427
428.
Laurent B, Peyron R, Garcia-Larrea L et al. (1999)
Inside the mechanisms of motor cortex
stimulation-induced analgesia. In 9th World
Congress on Pain, Book of Abstracts. Seattle, WA:
IASP Press, A187.
Lazorthes Y, Verdie JC, Arbus L (1978) Stimulation
analge`sique me`dullaire ante`rieure et poste`rieur
par technique dimplantation percutanee`. Acta
Neurochir (Wien) 40, 253
276.
Lazorthes Y (1979) European study on deep brain
stimulation. Resume`, Third European Workshop on
Electrical Neurostimulation. Paris: Medtronic.
Lazorthes Y, Siegfried J, Verdie JC, Casaux J (1995)
La stimulation me`dullaire chronique dans le
traitement des douleurs neuroge` nes. Etude
coope`rative et re`trospective sur 20 ans de suivi.
Neurochirurgie 41, 73
88.
Le Beau J, Daum S, Forjaz S (1948) Les tractotomies
trigeminales dans les traitement des nevralgies
faciales. Brasil Med Chir 10, 331
344.
Le Beau J, Choppy M, Gaches J, Rosier M (1954)
Psychochirurgie et fonctions mentales: techniques,
resultats, applications physiologiques. Paris:
Masson.
Le Pera D, Svensson P, Valeriani M et al. (2000)
Long-lasting effect evoked by tonic muscle pain on
parietal EEG activity in humans. Clin Neurophysiol
111, 2130
2137.
Lee MS, Choi IS, Chung TS (1989) Thalamic
syndrome and cortical hypoperfusion on technetium-99m HM-PAO brain SPECT. Yonsei Med J
30, 151
157.
Leijon G, Boivie J (1989a) Central post-stroke pain: a
controlled trial of amitriptyline and carbamazepine. Pain 36, 27
36.
Leijon G, Boivie J (1989b) Central post-stroke pain:
the effect of high and low frequency TENS. Pain
38, 187
191.
Leijon G, Boivie J, Johansson I (1989) Central
post-stroke pain: neurological symptoms and pain
characteristics. Pain 36, 13
25.
Lende RA, Kirsh WM, Druckman R (1971) Relief of
facial pain after combined removal of precentral
and postcentral cortex. J Neurosurg 34, 537
543.

358
Lenz FA (1991) The thalamus and central pain
syndromes: human and animal studies. In Casey
KL, ed., Pain and Central Nervous System Disease.
The Central Pain Syndromes. New York: Raven
Press, pp. 171
182.
Lenz FA, Kwan HC, Martin R et al. (1994)
Characteristics of somatotopic organization and
spontaneous neuronal activity in the region of the
thalamic principal sensory nucleus in patients
with spinal cord transections. J Neurophysiol 72,
1570
1587.
Lenz FA, Dougherty PM (1997) Pain processing in
the human thalamus. In Steriade M, Jones EG,
McCormick DA, eds., Thalamus: II. Experimental
and Clinical Aspects. Amsterdam: Elsevier,
pp. 617
652.
Lenz FA, Gracely RH, Baker FH, Richardson RT,
Dougherty PM (1998) Reorganization of sensory
modalities evoked by stimulation in the region of
the principal sensory nucleus (ventral caudal, Vc)
in patients with pain secondary to neural injury.
J Comp Neurol 399, 125
138.
Leriche R (1937) La chirurgie de la douleur. Paris:
Masson (3rd edn. 1949).
Levendoglu F, Ogun CO, Ozerbil O, Ogun TC,
Ugurlu H (2004) Gabapentin is a first line drug for
the treatment of neuropathic pain in spinal cord
injury. Spine 29, 743
751.
Levin AB (1988) Stereotactic chemical hypophysectomy. In Lunsford LD, ed., Modern Stereotactic Neurosurgery. Boston, MA: M Nijhoff,
pp. 365
374 (see also J Neurosurg 1983, 59,
1002
1006).
Levin G (1966) Electrical stimulation of the globus
pallidus and thalamus. J Neurosurg 24, 415.
Levy LM, Ziemann U, Chen R, Cohen LG (2002)
Rapid modulation of GABA in sensorimotor
cortex induced by acute deafferentation. Ann
Neurol 52, 755
761.
Levy RM, Lamb S, Adams JE (1987) Treatment
of chronic pain by deep brain stimulation: long
term follow-up and review of the literature.
Neurosurgery 21, 885
893.
Levy R, Lang AE, Dostrovsky JO et al. (2001)
Lidocaine and muscimol microinjections in subthalamic nucleus reverse Parkinsonian symptoms.
Brain 124, 2105
2118.
Lewin W, Philips CG (1952) Observations on partial
removal of the post-central gyrus for pain.
J Neurol Neurosurg Psychiat 15, 143
147.
Ley A (1957) Aneurismas arteriovenosos congenitos
intracraneales. Barcelona: Typografia la Academica.
Lhermitte J (1933) Physiologie des ganglions centraux. Les corps stries. La couche optique. Les

References
formations sousthalamiques. In: Binet R, ed.,
Traite` de physiologie normale et pathologique,
vol. 9. Paris: Masson, pp. 357
402.
Li J (2000) Clinical analysis and treatment of central
pain due to head injury. Chin J Traumatol 3,
126
127.
Libet B (1973) Electrical stimulation of cortex in
human subjects and conscious sensory aspects.
In Iggo A, ed., Handbook of Sensory Physiology,
vol. 2. Berlin: Springer, pp. 744
790.
Liepert J, Restemeyer C, Muenchau A, Weiller C
(2005) Motor cortex excitability after thalamic
infarction. Clin Neurophysiol 116, 1621
1627.
Linazasoro G (2004) Recent failures of new potential
symptomatic treatment for Parkinsons disease:
causes and solutions. Mov Disord 19, 743
754.
Lind G, Meyerson BA, Winter J, Linderoth B (2004)
Intrathecal baclofen as adjuvant therapy to
enhance the effect of spinal cord stimulation
in neuropathic pain: a pilot study. Eur J Pain 8,
377
383.
Lindblom U, Meyerson BA (1975) Influence on
touch, vibration and cutaneous pain of dorsal
column stimulation in man. Pain 1, 257
270.
Lindblom U (1991) New directions for basic and
clinical research in central pain syndromes. In
Casey KL, ed., Pain and Central Nervous System
Disease. The Central Pain Syndromes. New York:
Raven Press, pp. 275
280.
Lippert RS, Hosobuchi Y, Nielsen SL (1974) Spinal
commissurotomy. Surg Neurol 2, 373
377.
Lipton S (1989) Percutaneous cordotomy. In Wall
PD, Melzack R, eds., Textbook of Pain, 2nd edn.
Edinburgh: Churchill Livingstone, pp. 832
839.
Lisman J (1997) Bursts as a unit of neural
information: making unreliable synapses reliable.
TINS 20, 38
43.
Livingstone WK (1943) Pain Mechanisms: A Physiologic Interpretation of Causalgia and Its Related
States. New York: Macmillan (2nd edn: Plenum,
1946).
Llinas RR, Pare` D (1991) Commentary. Of dreaming
and wakefulness. Neuroscience 44, 521
535.
Llinas RR, Pare` D (1997) Coherent oscillations in
specific and nonspecific thalamocortical networks
and their role in cognition. In Steriade M,
Jones EG, McCormick DA, eds., Thalamus: II.
Experimental and Clinical Aspects. Amsterdam:
Elsevier, pp. 501
516.
Llinas RR, Ribary U, Jeanmonod D, Kronberg E,
Mitra PP (1999) Thalamocortical dysrhythmia:
a neurological and neuropsychiatric syndrome
characterized by magnetoencephalography.
Proc Natl Acad Sci USA 96, 15222
15227.

References
Loeser JD, Ward AA Jr., White LE Jr. (1968) Chronic
deafferentation of human spinal cord neurons.
J Neurosurg 29, 48
50.
Loh L, Nathan PW (1978) Painful peripheral states
and sympathetic blocks. J Neurol Neurosurg Psych
41, 664
671.
Loh L, Nathan PW, Schott GD, Wilson PG (1980)
Effects of regional guanethidine infusion in certain
painful states. J Neurol Neurosurg Psych 43,
446
451.
Loh L, Nathan PW, Schott GD (1981) Pain due to
lesions of central nervous system removed by
sympathetic block. Br Med J 282, 1026
1028.
Long DM, Erickson DE (1975) Stimulation of the
posterior columns of the spinal cord for relief of
intractable pain. Surg Neurol 4, 134
141.
Long DM, Hagfors N (1975) Electrical stimulation in
the nervous system: the current status of electrical
stimulation of the nervous system for relief of
pain. Pain 1, 109
123.
Long DM, Campbell JN, Gucer G (1979) Transcutaneous electrical stimulation for relief of chronic
pain. In Bonica JJ, Liebeskind JC, Albe-Fessard
DG, eds., Advances in Pain Research and Therapy.
New York: Raven Press, pp. 593
599.
Lorenz J, Kohlhoff H, Hansen H-C, Kunze, Bromm
B (1998) Abeta-fiber mediated activation of
cingulate cortex as correlate of central post-stroke
pain. NeuroReport 9, 659
663.
Loubser PG, Donovan WH (1991) Diagnostic spinal
anaesthesia in chronic spinal cord injury pain.
Paraplegia 29, 25
36.
Loubser PG, Clearman RR (1993) Evaluation of
central spinal cord injury pain with diagnostic
spinal anesthesia. Anesthesiology 79, 376
378.
Loubser PG, Akman NM (1996) Effects of intrathecal baclofen on chronic spinal cord injury pain.
J Pain Symptom Manage 12, 241
247.
Lozano AM, Parrent A, Tasker RR (1992) Central
pain from thalamic neoplasms. Stereotact Funct
Neurosurg 59, 77.
Luessenhop AJ, Dela Cruz T (1969) The surgical
excision of spinal intradural vascular malformation. J Neurosurg 30, 552
559.
Magnin M, Jetzer U, Morel A, Jeanmonod D (2001)
Microelectrode recording and macrostimulation
in thalamic and subthalamic MRI guided stereotactic surgery. Neurophysiol Clin 31, 230
238.
Mailis A, Amani N, Umana M, Basur R, Roe S
(1997) Effect of intravenous sodium amytal on
cutaneous sensory abnormalities, spontaneous
pain and algometric pain pressure thresholds in
neuropathic pain patients: a placebo-controlled
study. II. Pain 70, 69
81.

359
Mailis A, Bennett GJ (2002) Dissociation between
cutaneous and deep sensibility in central poststroke pain (CPSP). Pain 98, 331
334.
Maletic-Savatic M, Malinow R, Svoboda K (1999)
Rapid dendritic morphogenesis in CA1 hippocampal dendrites induced by synaptic activity.
Science 283, 1923
1927.
Mamie C, Morabia A, Bernstein M, Klopfenstein LE,
Forster A (2000) Treatment efficacy is not an index
of pain intensity. Can J Anaesth 47, 1166
1170.
Manduch M, Davis KD, Lozano AM, Tasker RR,
Dostrovsky JO (1999) Thalamic stimulationevoked pain and temperature sites in pain and
non-pain patients. In 9th World Congress on
Pain, Book of Abstracts. Seattle, WA: IASP Press,
A70-P151.
Mann L (1892) Kasuisticher Beitrag zur Lehre
vom Central entstehenden Schmerze. Berlin klin
Wochenschr 29, 244.
Mansuy L, Sindou M, Fischer G, Brunon J (1976) La
cordotomie spinothalamique dans les douleurs
cancereuses. Resultats dune serie de 124 malades
operes par abord direct posterieur. Neurochirurgie
22, 437
444.
Marchand S, Bushnell MC, Molina-Negro P,
Martinez SN, Duncan GN (1991) The effects of
dorsal column stimulation on measures of clinical
and experimental pain in man. Pain 45, 249
257.
Marchand S, Kupers RC, Bushnell MC, Duncan GH
(2003) Analgesic and placebo effects of thalamic
stimulation. Pain 105, 481
488.
Margot-Duclot A, Thiebaut J-B, Simon A et al.
(2002) Effects of intrathecal baclofen in cauda
equina and low spinal cord injury pain. In 10th
World Congress on Pain, Book of Abstracts. Seattle,
WA: IASP Press, A221-P217.
Marineo G (2003) Untreatable pain resulting from
abdominal cancer: new hope from biophysics?
JOP 4, 1
10.
Mark VH, Tsutsumi H (1974) The suppression of
pain by intrathalamic lidocaine. Adv Neurol 4,
715
721.
Mark VH, Ervin FR, Hackett TP (1960) Clinical
aspects of stereotactic thalamotomy in the human.
The treatment of chronic severe pain. Arch Neurol
3, 351
367.
Markram H, Toledo-Rodriguez M, Wang Y et al.
(2004) Interneurons of the neocortical inhibitory
system. Nature Rev Neurosci 5, 793
807.
Marshall J (1951) Sensory disturbances in cortical
wounds with special reference to pain. J Neurol
Neurosurg Psych 14, 187
204.
Massey EW (1984) Unilateral neurogenic pruritus
following stroke. Stroke 15, 901
903.

360
Masson C, Koskas P, Cambier J, Masson M (1991)
Syndrome cortical pseudothalamique gauche et
asymbolie a` la douleur. Rev Neurol (Paris) 147,
668
670.
Mauguiere F, Desmedt JE (1988) Thalamic pain
syndrome of Dejerine
Roussy: differentiation of
four subtypes assisted by somatosensory evoked
potentials data. Arch Neurol 45, 1312
1320.
Maurer M, Henn V, Dittrich A, Hofmann A (1990)
Delta-9-tetrahydrocannabinol shows antispastic
and analgesic effects in a single case doubleblind trial. Eur Arch Psychiatry Clin Neurosci
240, 1
4.
Max MB, Hagen NA (2000) Do changes in brain
sodium channels cause central pain? Neurology 54,
544
545.
Mayanagi Y, Bouchard G (1976
77) Evaluation of
stereotactic thalamotomies for pain relief, with
reference to pulvinar intervention. Appl Neurophysiol 39, 157
159.
Mayanagi Y, Sano K (1988) Posteromedial hypothalamotomy for behavioral disturbances and
intractable pain. In Lunsford LD, ed., Modern
Stereotactic Neurosurgery. Boston, MA: M Nijhoff,
pp. 377
384.
Mayanagi Y, Sano K (1998) Stimulation and
coagulation of the posterior hypothalamus for
intractable pain. In Gildenberg PL, Tasker RR,
eds., Textbook of Stereotactic and Functional
Neurosurgery. New York: McGraw-Hill,
pp. 1453
1454.
Mazars G, Pansini A, Chiarelli J (1960) Coagulation
du faisceau spinothalamique et du faiceau quintothalamique par stereotaxie. Indications-resultats.
Acta Neurochir 8, 324
326 (also includes J Neurol
Neurosurg Psych 1960, 23, 352).
Mazars GJ (1975) Intermittent stimulation of
nucleus ventralis posterolateralis for intractable
pain. Surg Neurol 4, 93
95 (also includes
Neurochirurgie 1974, 20, 117
124).
Mazars G (with Merienne L, Cioloca C) (1976) E`tat
actuel de la chirurgie de la douleur. Neurochirurgie
22(Suppl.), 1
164.
Mazars GJ, Merienne L, Cioloca C (1979) Comparative study of electrical stimulation of posterior
thalamic nuclei, periaqueductal gray, and other
midline mesencephalic stuctures in man. Adv Pain
Res Ther 3, 541
546.
McAlhany HS, Netsky MG (1955) Compression
of the spinal cord by extramedullary neoplasms.
J Neuropath Exp Neurol 14, 276
281.
McCance S, Hawton K, Brighouse D, Glynn C
(1996) Does electroconvulsive therapy (ECT)
have any role in the management of intractable

References
thalamic pain? Pain 68, 129
131.
McCarty GW (1954) The treatment of spastic paraplegia by selective spinal cordectomy. J Neurosurg
11, 539
545.
McCleane G (1998) Lamotrigine can reduce neurogenic pain associated with multiple sclerosis. Clin
J Pain 14, 268
270.
McCleane G (1998a) A prospective audit of the use
of lamotrigine in 300 chronic pain patients.
The Pain Clinic 11, 97
102.
McCormick PC, Torres R, Post KD, Stein BM (1990)
Intramedullary ependymoma of the spinal cord.
J Neurosurg 72, 523
533.
McCrone J (1999) Going Inside. Faber and Faber.
McGowan DJL, Janal MN, Clark WC et al. (1997)
Central post-stroke pain and Wallenbergs lateral
medullary infarction: frequency, character, and
determinants in 63 patients. Neurology 49,
120
125.
McQuay HJ, Carroll D, Jadad AR et al. (1994)
Dextrometorphan for the treatment of neuropathic pain: a double-blind randomised controlled
crossover trial with integral n-of-1 design. Pain 59,
127
133.
McQuay HJ, Tramer M, Nye BA et al. (1996) A
systematic review of antidepressants in neuropathic pain. Pain 68, 217
227.
Meador KJ, Ray PG, Day L, Ghelani H, Loring DW
(1998) Physiology of somatosensory perception:
cerebral lateralization and extinction. Neurology
51, 721
727.
Meador KJ, Ray PG, Echauz JR, Loring DW,
Vachtsevanos GJ (2002) Gamma coherence and
conscious perception. Neurology 24, 847
854.
Meadows J, Guarnieri M, Miller K et al. (2001) Type
I Chiari malformation: a review of the literature.
Neurosurg Quart 11, 220
229.
Meglio M (1998) Evaluation and management
of central and peripheral deafferentation pain.
In Gildenberg PL, Tasker RR, eds., Textbook
of Stereotactic and Functional Neurosurgery.
New York: McGraw-Hill, pp. 1631
1636.
Melzack R, Wall PD (1965) Pain mechanisms: a new
theory. Science 150, 971
979.
Melzack R, Loeser JD (1978) Phantom body
pain in paraplegics: evidence for a central
pattern generating mechanism for pain. Pain 4,
195
210.
Melzack R (1991) Central pain syndromes and
theories of pain. In Casey KL, ed., Pain and
Central Nervous System Disease. The Central Pain
Syndromes. New York: Raven Press, pp. 59
64.
Mercadante S (1998) Gabapentin in spinal cord
injury pain. Pain Clinic 10, 203
206.

References
Metzler J, Marks PS (1979) Functional changes in
cat somatic sensory-motor cortex short-term
during reversible epidural blocks. Brain Res 177,
379
383.
Merren MD (1998) Gabapentin for the treatment of
pain and tremor: a large case series. South Med J
91, 739
744.
Merskey H (1994) Classification of chronic pain:
descriptions of chronic pain syndromes and
definitions of pain terms. Pain Suppl. 3, 1986
(with Bogduk N, ed., 2nd edn. Seattle, WA: IASP
Press, 1994).
Meyer BU, Roericht S, Graefin von Einsiedel H,
Kruggel F, Weindl A (1995) Inhibitory and
excitatory interhemispheric transfers between
motor cortical areas in normal humans and
patients with abnormalities of the corpus callosum. Brain 118, 429
440.
Meyer-Lindenberg A, Ziemann U, Hajak G, Cohen L,
Berman KF (2002) Transitions between dynamical
states of differing stability in the human brain.
Proc Natl Acad Sci USA 99, 10948
10953.
Meyerson BA, Lindblom U, Linderoth B, Lind G,
Herregodts P (1993) Motor cortex stimulation
as treatment of trigeminal neuropathic pain.
Acta Neurochir 58(Suppl.), 150
153.
Meyerson BA, Linderoth B (1999) Electric stimulation of the central nervous system. In Max M, ed.,
Pain 1999: An Updated Review. Seattle, WA: IASP
Press, pp. 269
280.
Meyerson BA, Linderoth B (2001) Brain stimulation:
intracerebral and motor cortex stimulation. In
Loeser JD, ed., Bonicas Management of Pain,
3rd edn. Philadelphia, PA: Lippincott Williams &
Wilkins, pp. 1877
1889.
Michel D, Laurent B, Convers P et al. (1990),
Douleurs corticales. E`tude clinique, electrophysiologique et topographique de 12 cas. Rev Neurol
146, 405
414.
Michelsen JJ (1943) Subjective disturbances of the
sense of pain from lesions of the cerebral cortex.
Ass Res Nerv M Dis Proc 23, 86
99.
Middleton JW, Siddall PJ, Walker S, Molloy AR,
Rutkowski SB (1996) Intrathecal clonidine and
baclofen in the management of spasticity and
neuropathic pain following spinal cord injury: a
case study. Arch Phys Med Rehabil 77, 824
826.
Migita K, Uozumi T, Arita K, Monden S (1995)
Transcranial magnetic coil stimulation of motor
cortex in patients with central pain. Neurosurgery
36, 1037
1040.
Mikula F, Siroky J, Zapletal B (1959) Le traitement
des crises gastralgiques par la tractotomie mesencephalique bilaterale et ses complications auditives

361
inattendues. Rev oto-neuro oftal (Buenos Aires) 31,
456
463.
Milandre L, Brosset C, Khalil R (1993) Infarctus
thalamiques lateraux: 22 observations. Presse Med
22, 1865
1869.
Miles J (1998) The pituitary gland and pain relief.
In Gildenberg PL, Tasker RR, eds., Textbook of
Stereotactic and Functional Neurosurgery. New
York: McGraw-Hill, pp. 1457
1462.
Milhorat TH, Kotzen RM, Mu HTM, Capocelli AL,
Milhorat RH (1996) Dysesthetic pain in patients
with syringomyelia. Neurosurgery 38, 940
947.
Milhorat TH, Capocelli AL Jr., Kotzen RM et al.
(1997) Intramedullary pressure in syringomyelia:
clinical and pathophysiological correlates of syrinx
distension. Neurosurgery 41, 1102
1110.
Millan MJ (2002) Descending control of pain. Prog
Neurobiol 66, 355
474.
Miro` J, Garcia-Monco C, Leno C, Berciano C (1988)
Pelvic pain: an undescribed paroxysmal manifestation of multiple sclerosis. Pain 32, 73
75.
Miserocchi E (1951) Le cordotomie (tractotomie
spinali). Chirurgia 6, 519
538.
Mittal B, Thomas DG, Walton P, Calder I (1987)
Dorsal column stimulation in chronic pain: report
of 31 cases. Ann R Coll Surg Engl 69, 104
109.
Mladinich EK (1974) Diphenylhydantoin in the
Wallenberg syndrome. JAMA 230, 372
373.
Modesti LM, Waszak M (1975) Firing pattern of cells
in human thalamus during dorsal column stimulation. Appl Neurophysiol 38, 251
258.
Mogil JS (1999) Review. The genetic mediation
of individual differences in sensitivity to pain
and its inhibition. Proc Natl Acad Sci USA 96,
7745
7751.
Mogilner AY, Rezai AR (2001) Epidural motor
cortex stimulation with functional imaging guidance. Neurosurg Focus 11, article 4.
Mohammadian P, Hummel T, Loetsch J, Kobal G
(1997) Bilateral hyperalgesia to chemical stimulation of the nasal mucosa following unilateral
inflammation. Pain 73, 407
412.
Montes C, Mertens P, Convers P et al. (2002)
Cognitive effects of precentral cortical stimulation
for pain control: an ERP study. Neurophysiol Clin
32, 313
325.
Montes C, Magnin M, Maarrawi J et al. (2005)
Thalamic thermo-algesic transmission: ventral
posterior (VP) complex versus Vmpo in the light
of a thalamic infarct with central pain. Pain 113,
223
232.
Montgomery BM, King WW (1962) Hemiplegic
migraine; a case with paroxysmal shoulder
hand
syndrome. Ann Intern Med 57, 450
455.

362
Moore CEG, Schady W (2000) Investigation of
the functional correlates of reorganization within
the human somatosensory cortex. Brain 123,
1883
1895.
Moore CI, Nelson SB, Sur M (1999) Dynamics of
neuronal processing in rat somatosensory cortex.
TINS 22, 513
520.
Moore CI, Stern CE, Dunbar C et al. (2000) Referred
phantom sensations and cortical reorganization
after spinal cord injury in humans. Proc Natl Acad
Sci USA 97, 14703
14708.
Morgan M, Loh L, Singer L, Moore PH (1971)
Ketamine as the sole anesthetic agent for minor
surgical procedures. Anaesthesia 26, 158
159.
Mori S, Sadoshima S, Ibayashi S, Fujishima M, Iino
K (1995) Impact of thalamic hematoma on sixmonth mortality and motor and cognitive functional outcome. Stroke 26, 620
626.
Morin C, Bushnell MC, Luskin MB, Craig AD (2002)
Disruption of thermal perception in a multiple
sclerosis patient with central pain. Clin J Pain 18,
191
195.
Morley JS, Miles JB, Venn RF, Williams TSC (1991)
Dorsal horn enkephalins reduced in human
chronic pain. IASP Congress on Pain. Seattle,
WA: IASP Press, Abstr. 186.
Morley JS, Bridson J, Nash TP et al. (2003) Low-dose
methadone has an analgesic effect in neuropathic
pain: a double-blind randomized controlled crossover trial. Palliat Med 17, 576
587.
Morrow TJ, Casey KL (2002) Understanding central
pain: new insights from forebrain imaging studies
of patients and of animals with central lesions.
Progr Pain Res Manage 23, 265
279.
Moshe` SL (2000) Mechanisms of action of anticonvulsant agents. Neurology 55(Suppl. 1), S32
S40.
Moulin DE, Foley KM, Ebers GC (1988) Pain
syndromes in multiple sclerosis. Neurology 38,
1830
1834.
Mukherjee CS, Sarkhel A, Banerjee TK, Sen S (1999)
Community survey of central post-stroke pain
(CPSP). 9th World Congress on Pain, Book of
Abstracts. Seattle, WA: IASP Press, A338.
Mullan MJ (2002) Descending control of pain.
Progr Neurobiol 66, 355
474.
Muller F, Kunesch E, Binkofski F, Frend H-J (1991)
Residual sensorimotor functions in a patient after
right sided hemispherectomy. Neuropsychology 29,
125
145.
Mundinger F, Becker P (1977) Late results of central
stereotactic interventions for pain. Acta Neurochir
Suppl 24, 229.
Mundinger F, Salomao JF (1980) Deep brain
stimulation in mesencephalic lemniscus medialis

References
for chronic pain. Acta Neurochir Suppl 30,
245
258.
Mundinger F, Neumuller H (1982) Programmed
stimulation for control of chronic pain and motor
diseases. Appl Neurophysiol 45, 102
111.
Nagaro T, Amakawa K, Kimura S, Arai T (1993)
Reference of pain following percutaneous cervical
cordotomy. Pain 53, 205
211.
Nagaro T, Shimizu C, Inoue H et al. (1995) [The
efficacy of intravenous lidocaine on various types
of neuropathic pain]. Masui 44, 862
867.
Nair DR, Najm I, Bulacio J, Lueders H (2001)
Painful auras in focal epilepsy. Neurology 57,
700
702.
Nakazato Y, Yoshimaru K, Ohkuma A et al. (2004)
Central post-stroke pain in Wallenberg syndrome.
No To Shinkei 56, 385
388.
Namba S, Nakao Y, Matsumoto Y, Ohmoto T,
Nishimoto A (1984) Electrical stimulation of
the posterior limb of the internal capsule for
treatment of thalamic pain. Appl Neurophysiol
47, 137
148 (see also J Neurosurg 1985, 63,
224
234).
Nandi D, Smith H, Owen S et al. (2002) Periventricular grey stimulation versus motor cortex
stimulation for post-stroke neuropathic pain.
J Clin Neurosci 9, 557
561.
Nandi D, Aziz T, Carter H, Stein J (2003) Thalamic
field potentials in chronic central pain treated by
periventricular gray stimulation
a series of eight
cases. Pain 101, 97
107.
Nandi D, Aziz TZ (2004) Deep brain stimulation in
the management of neuropathic pain and multiple
sclerosis tremor. J Clin Neurophysiol 21, 31
39.
Nashold BS, Wilson WP, Slaughter DG (1969)
Stereotactic midbrain lesions for central dysesthesia and phantom pain. Preliminary report.
J Neurosurg 30, 116
126.
Nashold BS, Wilson WP (1970) Central pain and
irritable midbrain. In Crue B, ed., Pain and
Suffering. Selected Aspects. Springfield, IL: CC
Thomas, pp. 95
118 (also Conf Neurol 1966, 27,
30
34; Adv Neurol 1974, 191
196).
Nashold BS, Friedman H Jr. (1972) Dorsal column
stimulation for control of pain. Preliminary report
on 30 patients. J Neurosurg 30, 590
597.
Nashold BS Jr. (1974) Central pain: its origin and
treatment. Clin Neurosurg 21, 311
322.
Nashold BS Jr. (1975) Dorsal column stimulation
for control of pain: a three-year follow-up. Surg
Neurol 4, 146
147.
Nashold BS Jr., Bullit E (1981) Dorsal root entry
zone lesions to control central pain in paraplegics.
J Neurosurg 55, 414
419.

References
Nashold BS Jr. (1982) Brainstem stereotaxic procedures. In Schaltenbrand G, Walker AE, eds.,
Stereotaxy of the Human Brain. Anatomical,
Physiological and Clinical Applications, 2nd edn.
Stuttgart: Georg Thieme Verlag, pp. 475
483.
Nashold BS (1984) Current status of the DREZ
operation. Neurosurgery 15, 942
944.
Nashold BS Jr. (1991) Paraplegia and pain. Adv Pain
Res Ther 19, 301.
Nashold BS, Pearlstein RD, eds. (1996) The DREZ
Operation. Park Ridge, IL: AANS Publications.
Nasreddine ZS, Saver JL (1997) Pain after thalamic
stroke: right diencephalic predominance and
clinical features in 180 patients. Neurology 48,
1196
1199.
Nasreddine ZS, Saver JL (1998) Pain after thalamic
stroke: right diencephalic predominance and
clinical features in 180 patients. Neurology 51,
927
928.
Nathan PW (1960) Improvement in cutaneous
sensibility associated with relief of pain. J Neurol
Neurosurg Psych 23, 202
206.
Nathan PW, Smith MC (1972) Pain in cancer:
comparison of results of cordotomy and chemical
rhizotomy. In Fusek I, Kunc Z, eds., Present Limits
of Neurosurgery. Amsterdam: Excerpta Medica,
pp. 513
519.
Nathan PW, Smith MC (1979) Clinico-anatomical
correlation in anterolateral cordotomy. In Bonica
JJ, Liebeskind JC, Albe-Fessard DG, Jones LE, eds.,
Advances in Pain Research and Therapy. New York:
Raven Press, pp. 921
926.
Nathan PW, Smith MC (1984) Dysesthe`sie apre`s
cordotomie. Med Hygiene 42, 1788
1790.
Nathan PW, Smith MC, Cook AW (1986) Sensory
effects in man of lesions of the posterior columns
and of some other afferent pathways. Brain 109,
1003
1041.
Nathanson M (1962) Paroxysmal phenomena
resembling seizures related to spinal cord and
root pathology. J Mt Sinai Hosp 29, 147
151.
Naver H, Blomstrand C, Ekholm S et al. (1995)
Autonomic and thermal sensory symptoms
and dysfunction after stroke. Stroke 26,
1379
1385.
Nelson SB, Turrigiano GG (1998) Synaptic depression: a key player in the cortical balancing act.
Nature Neurosci 1, 539
541.
Nepomuceno C, Fine PR, Richards JS et al. (1979)
Pain in patients with spinal cord injury. Arch Phys
Med Rehab 60, 605
609.
Ness TJ, San Pedro EC, Richards JS et al. (1998)
Clinical note. A case of spinal cord injury-related
pain with baseline rCBF brain SPECT imaging

363
and beneficial response to gabapentin. Pain 78,
139
143.
Ness TJ, Putzke JD, Liu H-G, Mountz J (2002)
Examples of the use of gabapentin in the
treatment of spinal cord injury pain. Progr Pain
Res Manage 23, 379
392.
Nguyen DK, Botez MI (1998) Diaschisis and
neurobehavior. Can J Neurol Sci 25, 5
12.
Nguyen JP, Keravel Y, Feve A et al. (1997) Treatment
of deafferentation pain by chronic stimulation of
the motor cortex: report of a series of 20 cases.
Acta Neurochir (Wien) Suppl 68, 54
60.
Nguyen JP, Lefacheur JP, Decq P et al. (1999)
Chronic motor cortex stimulation in the treatment of central and neuropathic pain: correlations
between clinical, electrophysiological and anatomical data. Pain 82, 245
251.
Nicholson K, Mailis A, Taylor A (2002) Psychometric correlates of placebo and nocebo responses.
In 10th World Congress on Pain, Book of Abstracts.
Seattle, WA: IASP Press, Abst. 317-P596.
Niizuma H, Kwak R, Ikeda S et al. (1982) Follow-up
results of centromedian thalamotomy for central
pain. Appl Neurophysiol 45, 324
325.
Nitescu P, Dahm P, Appelgren L, Curelaru I (1998)
Continuous infusion of opioid and bupivacaine
by externalized intrathecal catheters in long-term
treatment of refractory nonmalignant pain.
Clin J Pain 14, 17
28.
Noordenbos W (1959) Pain, Problems Pertaining
to the Transmission of Nerve Impulses Which Give
Rise to Pain. Amsterdam: Elsevier, p. 135.
Noordenbos W, Wall PD (1976) Diverse sensory
functions with an almost totally divided spinal
cord. A case of spinal cord transection with
preservation of part of one anterolateral quadrant.
Pain 2, 185
195.
North RB, Kidd DH, Zahurak M, James CS, Long
DM (1993) Spinal cord stimulation for chronic,
intractable pain: experience over two decades.
Neurosurgery 32, 384
395.
Notcutt W, Price M, Miller R et al. (2004) Initial
experience with medicinal extracts of cannabis
for chronic pain: results from 34 N of 1 studies.
Anaesthesia 59, 440.
Nowak LG, Bullier J (1998) Axons, but not cell
bodies, are activated by electrical stimulation
in cortical gray matter: I. Evidence from
chronaxie measurements; II. Evidence from
selective inactivation of cell bodies and axon
initial segments. Exp Brain Res 118, 477
488;
489
500.
Nusser Z, Hajos N, Somogyi P, Mody I (1998)
Increased number of synaptic GABA(A) receptors

364
underlies potentiation at hippocampal inhibitory
synapses. Nature 395, 172
177.
Nuti C, Peyron R, Garcia-Larrea L et al. (2005)
Motor cortex stimulation for refractory
neuropathic pain: four year outcome and predictors of efficacy. Pain 118, 43
52.
Nyquist JK, Greenhoot JH (1973) Responses
evoked from the thalamic centrum medianum by
painful input: suppression by dorsal funiculus
conditioning. Exp Neurol 39, 215
222.
Oaklander AL, Romans K, Horasek S et al. (1998)
Unilateral postherpetic neuralgia is associated
with bilateral sensory neuron damage. Ann Neurol
44, 789
795.
Obersteiner H (1881) On allochiria. A peculiar
sensory disorder. Brain 4, 153
163.
Obeso JA, Rodriguez-Oroz MC, Rodriguez M et al.
(2000) Pathophysiologic basis of surgery for
Parkinsons disease. Neurology 55(Suppl. 6),
S7
S12.
Obrador S (1956) Quoted by Laine and Gros (1956).
Obrador S, Dierssen G, Ceballos R (1957) Consideraciones clinicas, neurologicas y anatomicas
sobre el llamado dolor talamico. Con motivo de
dos casos personales. Acta Neurol Latinoamer 3,
58
77.
Obrador S, Carrascosa R, Sevillano M (1961)
Observaciones sobre la estimulac`on y lesio`n
tala`mica en mano fantasma dolorosa. Rev esp
Oto-Neuro-Oftal 20, 149
153.
Ochoa JL (1993) The human sensory unit and pain:
new concepts, syndromes, and tests. Muscle Nerve
16, 1009
1016.
Ochoa JL (1999) Truths, errors, and lies around
reflex sympathetic dystrophy and complex
regional pain syndrome. J Neurol 246, 875
879.
Ohta Y, Akino M, Iwasaki Y, Abe H (1992) Spinal
epidural stimulation for central pain caused by
a cord lesion. No Shinkei Geka 20, 147
152.
Ohye C, Narabayashi H (1972) Activity of thalamic
neurons and their receptive fields in different
functional states in man. In Somjen GG, ed.,
Neurophysiology Studies in Man. Amsterdam:
Excerpta Medica, pp. 79
84.
Ohye C (1990) Thalamus. In Paxinos G, ed., The
Human Nervous System. San Diego, CA: Academic
Press, pp. 439
468.
Ohye C (1998) Stereotactic treatment of central pain.
Stereotact Funct Neurosurg 70, 71
76.
Ojemann JG, Silbergeld DL (1995) Cortical stimulation mapping of phantom limb rolandic cortex.
Case report. J Neurosurg 82, 641
644.
Oke A, Keller R, Meffond I, Adams RN (1978)
Lateralization of norepinephrine in the human

References
thalamus. Science 200, 1411
1412.
Olausson H, Marchand S, Bittar RG et al. (2001a)
Central pain in a hemispherectomized patient.
Eur J Pain 5, 209
217.
Olausson H, Ha B, Duncan GH et al. (2001b)
Cortical activation by tactile and painful stimuli
in hemispherectomized patients. Brain 124,
916
927.
Olivecrona U (1947) The surgery of pain. Acta Psych
Scand Suppl 46, 268
280.
Oliveira RA, Teixeira MJ (2002) Central poststroke
pain: clinical
encephalic imaging correlations.
In 10th World Congress on Pain, Book of Abstracts.
Seattle, WA: IASP Press, A96-P92.
Olsen RW, Avoli M (1997) GABA and epileptogenesis. Epilepsia 38, 399
407.
Olsen RW, Betz H (2006) GABA and Glycine. In:
Siegel GJ, ed., Basic Neurochemistry. Amsterdam:
Elsevier Academic Press, 291
302.
Orthner H, Roeder F (1966) Further clinical
and anatomical experiences with stereotactic
operations for relief of pain. Confin Neurol 27,
418
430.
Osterberg A, Boivie J, Holmgren H, Thuomas K,
Johansson I (1994) The clinical characteristics
and sensory abnormalities of patients with central
pain caused by multiple sclerosis. In Gebhart GF,
Hammond DL, Jensen TS, eds., Proceedings of the
7th World Congress on Pain. Seattle, WA: IASP
Press, pp. 789
796.
Osterberg A, Boivie J, Thuomas KA (2005) Central
pain in multiple sclerosis: prevalence and clinical
characteristics. Eur J Pain 9, 531
542.
Owen SLF, Green AL, Stein JF, Aziz TZ (2006)
Deep brain stimulation for the alleviation of poststroke neuropathic pain. Pain 120, 202
206.
Paciaroni M, Bogousslavsky J (1998) Pure sensory
syndromes in thalamic stroke. Eur Neurol 39,
211
217.
Pagni CA (1977) Central pain and painful anesthesia.
Pathophysiology and treatment of sensory deprivation syndromes due to central and peripheral
nervous system lesions. Progr Neurol Surg 8,
132
257.
Pagni CA, Regolo P (1987) Epilessia sensitivomotoria di origine spinale. Sintomo di esordio
raro dei meningiomi spinali. Minerva Chir 42,
2003
2009.
Pagni CA (1989) Central pain due to spinal cord and
brain stem damage. In Wall PD, Melzack R, eds.,
Textbook of Pain, 2nd edn. Edinburgh: Churchill
Livingstone, pp. 634
655.
Pagni CA, Canavero S (1993) Paroxysmal perineal
pain resembling tic douloureux, only symptom

References
of a dorsal meningioma. Ital J Neurol Sci 14,
323
324.
Pagni CA, Lanotte M, Canavero S (1993) How
frequent is anesthesia dolorosa following spinal
posterior rhizotomy? A retrospective analysis of
fifteen patients. Pain 54, 323
327.
Pagni CA, Canavero S (1995a) Functional thalamic
depression in a case of reversible central pain
due to a spinal intramedullary cyst. Case report.
J Neurosurg 83, 163
165.
Pagni CA, Canavero S (1995b) Cordomyelotomy in
the treatment of paraplegia pain. Experience in
two cases with long-term results. Acta Neurol Belg
95, 33
36.
Panerai AE, Monza G, Movilia P et al. (1990) A
randomized, within-patient, cross-over, placebocontrolled trial on the efficacy and tolerability of
the tricyclic antidepressants chlorimipramine and
nortriptyline in central pain. Acta Neurol Scand
82, 34
38.
Pareti G, De Palma A (2004) Does the brain
oscillate? The dispute on neuronal synchronization. Neurol Sci 25, 41
47.
Parker HL (1930) Pain of central origin. Am J Med
Sci 179, 241
258.
Parrent AG, Tasker RR (1992) Can the ipsilateral
hemisphere mediate pain in man? Acta Neurochir
(Wien) 117, 89.
Parrent AG, Lozano AM, Dostrovsky JO, Tasker RR
(1992) Central pain in the absence of functional
sensory thalamus. Stereotact Funct Neurosurg
59, 9
14.
Pattany PM, Widerstroem-Noga EG, Bowen BC
et al. (2002) Proton magnetic resonance spectroscopy following spinal cord injury: evaluation of
patients with chronic neuropathic pain. Progr
Brain Res Manage 23, 301
311.
Penfield W, Gage L (1933) Cerebral localization of
epileptic manifestations. Arch Neurol Psychiat 30,
709
727.
Penfield W, Welch K (1951) The supplementary
motor area of the cerebral cortex: a clinical and
experimental study. Arch Neurol Psych 66,
298
317.
Penfield W, Jasper H (1954) Epilepsy and the
Functional Anatomy of the Human Brain. Boston,
MA: Little, Brown.
Penn RD, Paice JA (2000) Adverse effects associated
with the intrathecal administration of ziconotide.
Pain 85, 291
296.
Percheron G (2004) Thalamus. In Paxinos G, Mai
JK, eds., The Human Nervous System, 2nd edn.
Amsterdam: Elsevier Academic Press,
pp. 592
676.

365
Petit-Dutaillis D, Messimy R, Feld H (1950)
Troubles sensitifs et sensorielles apres ablation
prefrontal chez lhomme. Rev Neurol 83, 23.
Petit-Dutaillis D, Messimy R, Berges L (1953) La
psychochirurgie des algies irreductibles. Etude
basee sur 57 cas. Semaine des Hopitaux de Paris
29, 3893
3903.
Petrovic P, Ingvar M (2002) Imaging cognitive
modulation of pain processing. Pain 95, 1
5.
Peyron R, Garcia-Larrea L, Deiber MP et al. (1995)
Electrical stimulation of precentral cortical area in
the treatment of central pain: electrophysiological
and PET study. Pain 62, 275
286.
Peyron R, Garcia-Larrea L, Gregoire MC et al. (1998)
Allodynia after lateral-medullary (Wallenberg)
infarct. A PET study. Brain 121, 345
356.
Peyron R, Garcia-Larrea L, Costes N et al. (1999)
The haemodynamic pattern of central pain
patients with allodynia. In 9th World Congress
on Pain, Book of Abstracts. Seattle, WA: IASP
Press, A78.
Peyron R, Garcia-Larrea L, Gregoire MC et al. (2000)
Parietal and cingulate processes in central
pain. A combined positron emission tomography
(PET) and functional magnetic resonance
imaging (fMRI) study of an unusual case. Pain
84, 77
87.
Peyron R, Schneider F, Faillenot I et al. (2004) An
fMRI study of cortical representation of mechanical allodynia in patients with neuropathic pain.
Neurology 63, 1838
1846.
Pfeiffer A, Pasi A, Mehraein P, Herz A (1982) Opiate
binding sites in human brain. Brain Res 248,
87
96.
Phillips NI, Bhakta BB (2000) Effect of deep brain
stimulation on limb paresis after stroke. Lancet
356, 222
223.
Pillay PK, Hassenbusch SJ (1992) Bilateral MRIguided stereotactic cingulotomy for intractable
pain. Stereotact Funct Neurosurg 59, 33
38.
Pioli EY, Gross CE, Meissner W, Bioulac BH, Bezard
E (2003) The deafferented nonhuman primate is
not a reliable model of intractable pain. Neurol Res
25, 127
129.
Pirotte B, Voordecker P, Neugroschl C et al. (2005)
Combination of functional magnetic resonance
imaging-guided neuronavigation and intraoperative cortical brain mapping improves targeting of
motor cortex stimulation in neuropathic pain.
Neurosurgery 56(ONS Suppl. 2), 344
359.
Ploner M, Gross J, Timmermann L, Schnitzler A
(2002) Cortical representation of first and second
pain sensation in humans. Proc Natl Acad Sci USA
99, 12444
12448.

366
Plotkin R (1982) Results in 60 cases of deep brain
stimulation for chronic intractable pain. Appl
Neurophysiol 45, 173
178.
Pollo A, Amanzio M, Arslanian A et al. (2001)
Response expectancies in placebo analgesia and
their clinical relevance. Pain 93, 77
84.
Pollock LJ, Boshes B, Finkelmann I et al. (1951a)
Management of residual injuries to spinal cord
and cauda equina. JAMA 146, 1551
1563.
Pollock LJ, Brown M, Boshes B et al. (1951b) Pain
below the level of injury of the spinal cord. Arch
Neurol Psychiatr 65, 319
322.
Pool JL (1946) Posterior cordotomy for relief of
phantom limb pain. Ann Surg 124, 386
391.
Porreca F, Ossipov MH, Gebhart GF (2002) Review.
Chronic pain and medullary descending facilitation. TINS 25, 319
325.
Portenoy RK, Foley KM (1986) Chronic use of
opioid analgesics in non-malignant pain. Report
of 38 cases. Pain 25, 171
186.
Portenoy RK, Yang K, Thornton D (1988) Chronic
intractable pain: an atypical presentation of
multiple sclerosis. J Neurol 235, 226
228.
Portenoy RK, Foley KM, Inturrisi CE (1990) The
nature of opioid responsiveness and its implications for neuropathic pain: new hypotheses
derived from studies of opioid infusions. Pain
43, 273
286.
Porter RW, Hohmann GW, Bors E, French JD
(1966) Cordotomy for pain following cauda
equina injury. Arch Surg 92, 765
770.
Potagas C, Avdelidis D, Singounas E, Missir O,
Aessopos A (1997) Episodic pain associated with
a tumor in the parietal operculum: a case report
and literature review. Pain 72, 201
208.
Potter PJ, Hayes KC, Hsieh JT, Delaney GA, Segal
JL (1998) Sustained improvements in neurological function in spinal cord injured patients
treated with oral 4-aminopyridine: three cases.
Spinal Cord 36, 147
155.
Pound P, Ebrahim S, Sandercock P, Bracken MB,
Roberts I on behalf of the Reviewing Animal Trials
Systematically (RATS) Group (2004) Where is the
evidence that animal research benefits humans?
Br Med J 328, 514
517.
Powers SK, Barbaro NM, Levy RM (1988) Pain
control with laser-produced dorsal root entry zone
lesions. Appl Neurophysiol 51, 243
254 (see also
J Neurosurg 1984, 61, 841
847).
Preissl H, Flor H, Lutzenberger W et al. (2001) Early
activation of the primary somatosensory cortex
without conscious awareness of somatosensory
stimuli in tumor patients. Neurosci Lett 308,
193
196.

References
Prestor B (2001) Microsurgical junctional DREZ
coagulation for treatment of deafferentation pain
syndromes. Surg Neurol 56, 259
265.
Procacci P, Maresca M (1991) Central pruritus.
Case report. Pain 47, 369
370.
Putzke JD, Richards JS, Kezar L, Hicken BL, Ness TJ
(2002) Long-term use of gabapentin for treatment
of pain after traumatic spinal cord injury. Clin J
Pain 18, 116
121.
Quarti M, Terzian H (1954) Lemisferectomia nel
trattamento delle emiplegie infantili. Chirurgia 9,
339
350.
Quigley DG, Arnold J, Eldridge PR et al. (2003)
Long-term outcome of spinal cord stimulation
and hardware complications. Stereotact Funct
Neurosurg 81, 50
56.
Rabben T, Skjelbred P, Oye I (1999) Prolonged analgesic effect of ketamine, an N-methyl-D-aspartate
receptor inhibitor, in patients with chronic pain.
J Pharmacol Exp Ther 289, 1060
1066.
Radhakrishnan V, Tsoukatos J, Davis KD et al.
(1999) A comparison of the burst activity of lateral
thalamic neurons in chronic pain and non-pain
patients. Pain 80, 567
575.
Raichle ME, McLeod AM, Snyder AZ et al. (2001)
A default mode of brain function. Proc Natl Acad
Sci USA 98, 676
682.
Raij TT, Numminen J, Narvanen S, Hiltunen J,
Hari R (2005) Brain correlates of subjective reality
of physically and psychologically induced pain.
Proc Natl Acad Sci USA 102, 2147
2151.
Ranck JB (1975) Which elements are excited in
electrical stimulation of mammalian central nervous system: a review. Brain Res 98, 417
440.
Rasche D, Ruppolt M, Stippich C, Unterberg A,
Tronnier VM (2006) Motor cortex stimulation
for long-term relief of chronic neuropathic pain:
a 10 year experience. Pain 121, 43
52.
Rasmisky M (1980) Physiology of conduction in
demyelinated axons. In Waxman, SG ed., Physiology and Pathobiology of Axons. New York: Raven
Press, pp. 361
376.
Rasmussen PV, Sindrup SH, Jensen TS, Bach FW
(2004) Therapeutic outcome in neuropathic pain:
relationship to evidence of nervous system lesion.
Eur J Neurol 11, 545
553.
Rath SA, Braun V, Soliman N, Antoniadis G, Richter
MP (1996) Results of DREZ coagulations for pain
related to plexus lesions, spinal cord injuries and
postherpetic neuralgia. Acta Neurochir (Wien)
138, 364
369.
Ray CD, Burton CV (1980) Deep brain stimulation
for severe chronic pain. Acta Neurochir (Wien)
Suppl 30, 289
293.

References
Reboiledo PE, Gonzalez X, Valenzuela P, Larraguibel
F, Mujica A (2002) Study of gabapentin and
clomipramine use for chronic pain management
in spinal cord injury. In 10th World Congress on
Pain, Book of Abstracts. Seattle, WA: IASP Press,
A925-P195.
Regev I, Avrahami A, Bornstein N, Koreczyn AD
(1983) Pain and hyperpathia in pontine haemorrhage. J Neurol 230, 205
208.
Reig E (1993) Spinal infusion of morphine for the
treatment of neuropathic pain. Medtronic Conference on Advances in Chronic Pain Treatment,
Padova, November 27, 1993.
Remy P, Zilbovicius M, Cesaro P et al. (1999)
Primary somatosensory cortex activation is not
altered in patients with ventroposterior thalamic
lesions: a PET study. Stroke 30, 2651
2658.
Retif J (1963) Douleur centrale et lesion suprathalamique. Meningiome temporo-parietal se manifestant par un syndrome algique paroxystique
a` caracte`re pseudo-radiculaire du membre inferieur controlateral. Acta Neurol Psychiat Belg 63,
955
969.
Retif J, Brihaye J, Vanderhaegen JJ (1967) Syndrome
douloureux thalamique et lesion parietale. A`
propos de trois observations de tumeur a`
localisation parietale, setant accompagnees de
douleurs spontanees de lhemicorps contralateral.
Neurochirurgie 13, 375
384.
Reynolds DV (1969) Surgery in the rat during
electrical analgesia induced by focal brain stimulation. Science 164, 444
445.
Rezai AR, Lozano AM, Crawley AP et al. (1999)
Thalamic stimulation and functional magnetic
resonance imaging: localization of cortical
and subcortical activation with implanted
electrodes. Technical note. J Neurosurg 90,
583
590.
Richardson DE (1974) Thalamotomy for control of
chronic pain. Acta Neurochir (Wien) Suppl 21,
77
88.
Richardson DE, Akil H (1977a) Pain reduction by
electrical brain stimulation in man: 1. Acute
administration in periaqueductal and periventricular sites; 2. Chronic self-administration in the
periventricular gray matter. J Neurosurg 47,
178
183; 184
194.
Richardson DE, Akil H (1977b) Long-term results of
periventricular gray self-stimulation. Neurosurgery
1, 199
202.
Richardson RR, Meyer PR, Cerullo LJ (1980)
Neurostimulation in the modulation of intractable
paraplegic and traumatic neuroma pains. Pain 8,
76
84.

367
Richter HP, Seitz K (1984) Dorsal root entry zone
lesions for the control of deafferentation pain:
experiences in ten patients. Neurosurgery 15,
956
959.
Riddoch G, Critchley McD (1937) La physiopathologie de la douleur dorigine centrale. Rev Neurol
68, 77
104.
Riddoch G (1938) The clinical features of central
pain. Lumleian lecture. Lancet 234, 1093
1098;
1150
1156; 1205
1209.
Ridgeway B, Wallace M, Gerayli A (2000) Ziconotide
for the treatment of severe spasticity after spinal
cord injury. Pain 85, 287
289.
Riechert T (1961) Methodes stereotaxiques dans la
chirurgie de la douleur. Exc Medica Int Congr Ser
36, 34
35f.
Rinaldi PC, Young RF, Albe-Fessard D,
Chodakiewitz J (1991) Spontaneous neuronal
hyperactivity in the medial and intralaminar
thalamic nuclei of patients with deafferentation
pain. J Neurosurg 74, 415
421.
Rodriguez RF, Contreras N (2002) Bilateral motor
cortex stimulation for the relief of central
dysesthetic pain and intentional tremor secondary
to spinal cord surgery: a case report. Neuromodulation 5, 189
195.
Roeder F, Orthner H (1961) Erfahrungen mit
ber
sterotaktischen Eingriffen. III Mitteilung. U
zerebrale Schmerzoperationen, insbesondere mediale Mesencephalotomie bei thalamischer Hyperpathie und bei Anaesthesia Dolorosa. Confin
Neurol 21, 51
97.
Rog DJ, Nurmikko TJ, Friede T, Young CA (2005)
Randomized, controlled trial of cannabis based
medicine in central pain in multiple sclerosis.
Neurology 65, 812
819.
Rogano L, Teixeira MJ, Lepski G (2003) Chronic
pain after spinal cord injury: clinical characteristics. Stereotact Funct Neurosurg 81, 65
69.
Rogers R, Painter D, Wise RG, Longe SE, Tracey I
(2002) Investigating ketamine analgesia in
humans using functional magnetic resonance
imaging. In 10th World Congress on Pain, Book of
Abstracts. Seattle, WA: IASP Press, A1123-P39.
Romanelli P, Heit G (2004) Patient-controlled
deep brain stimulation can overcome analgesic tolerance. Stereotact Funct Neurosurg 82,
77
79.
Rosen JA, Barsoum AH (1979) Failure of chronic
dorsal column stimulation in multiple sclerosis.
Ann Neurol 6, 66
67.
Rosenberg JM, Harrell C, Ristic H, Werner RA, de
Rosayro AM (1997) The effect of gabapentin on
neuropathic pain. Clin J Pain 13, 251
255.

368
Rosner H, Rubin L, Kestenbaum A (1996) Gabapentin adjunctive therapy in neuropathic pain
states. Clin J Pain 12, 56
58.
Rosomoff HL (1969) Bilateral percutaneous cervical
radiofrequency cordotomy. J Neurosurg 31,
41
46.
Rossetti AO, Ghika JA, Vingerhoets F, Novy J,
Bogousslavsky J (2003) Neurogenic pain and
abnormal movements contralateral to an anterior
parietal artery stroke. Arch Neurol 60, 1004
1006.
Rosso T, Aglioti SM, Zanette G et al. (2003)
Functional plasticity in the human primary
somatosensory cortex following acute lesion of
the anterior lateral spinal cord: neurophysiological
evidence of short-term cross-modal plasticity.
Pain 101, 117
127.
Rousseaux M, Cassim F, Bayle B, Laureau E (1999)
Analysis of the perception of and reactivity to pain
and heat in patients with Wallenberg syndrome
and severe spinothalamic tract dysfunction. Stroke
30, 2223
2229.
Roussy G (1906) Les couches optiques: etude
anatomique, physiologique et clinique. These de
Paris.
Roxin A, Riecke H, Solla SA (2004) Self-sustained
activity in a small-world network of excitable
neurons. Phys Rev Lett 92, 198101.
Rowbotham GF (1961) A case of intractable pain
in the head and face associated with pathological
changes in the optic thalamus. Acta Neurochir 9,
1
18.
Rowbotham MC, Twilling L, Davies PS et al. (2003)
Oral opioid therapy for chronic peripheral and
central neuropathic pain. New Engl J Med 348,
1223
1232.
Rudolph U, Antkowiak B (2004) Molecular and
neuronal substrates for general anaesthetics.
Nature Rev Neurosci 5, 709
720.
Sackeim HA, Prudic J, Devanand DP et al. (1993)
Effects of stimulus intensity and electrode placement on the efficacy and cognitive effects of
electroconvulsive therapy. New Engl J Med 328,
839
846.
Sadzot B, Mayberg HS, Frost JJ (1990) Detection and
quantification of opiate receptors in man by
positron emission tomography: potential applications to the study of pain. Neurophysiol Clin 20,
323
334.
Sage JI (2004) Pain in Parkinsons disease. Curr Treat
Options Neurol 6, 191
200.
Saitoh Y, Kato A, Ninomiya H et al. (2000) Primary
motor cortex stimulation within the central sulcus
for treating deafferentation pain. Acta Neurochir
Suppl 87, 149
152.

References
Saitoh Y, Osaki Y, Nishimura H et al. (2004)
Increased regional cerebral blood flow in the
contralateral thalamus after successful motor
cortex stimulation in a patient with poststroke
pain. J Neurosurg 100, 935
939.
Sakai T, Tomiyasu S, Ono T, Yamada H,
Sumikawa K (2004) Multiple sclerosis with severe
pain and allodynia alleviated by oral ketamine.
Clin J Pain 20, 375
376.
Sakurai M, Kanazawa I (1999) Positive symptoms
in multiple sclerosis: their treatment with sodium
channel blockers, lidocaine and mexiletine.
J Neurol Sci 162, 162
168.
Salin PA, Prince DA (1996) I. Spontaneous GABA-a
receptor-mediated inhibitory currents in adult rat
somatosensory cortex; II. Electrophysiological
mapping of GABA-A receptor-mediated inhibition in adult rat somatosensory cortex. J Neurophysiol 75, 1573
1588; 1589
1600.
Salmon JB, Hanna MH, Williams M, Toone B,
Wheeler M (1988) Thalamic pain: the effect of
electroconvulsive therapy. Pain 33, 67
71.
Samii M, Moringlane JR (1984) Thermocoagulation
of the dorsal root entry zone for the treatment of
intractable pain. Neurosurgery 15, 953
955.
Sampson JH, Nashold BS Jr. (1992) Facial pain due
to vascular lesions of the brain stem relieved by
dorsal root entry zone lesions in the nucleus
caudalis. Report of two cases. J Neurosurg 77,
473
475.
Samuelsson M, Samuelsson L, Lindell D (1994)
Sensory symptoms and signs and results of
quantitative sensory thermal testing in patients
with lacunar infarct syndromes. Stroke 25,
2165
2170.
Sanchez-Vives MV, McCormick DA (2000) Cellular
and network mechanisms of rhythmic recurrent activity in neocortex. Nature Neurosci 3,
1027
1034.
Sandkuehler J (1996) The organization and function
of endogenous antinociceptive systems. Progr
Neurobiol 50, 49
81.
Sandroni P (2002) Central neuropathic itch: a new
treatment option? Neurology 59, 778
779.
Sandyk R (1985) Spontaneous pain, hyperpathia
and wasting of the hand due to parietal lobe
haemorrhage. Eur Neurol 24, 1
3.
Sanford PR, Lindblom LB, Haddox JD (1992)
Amitriptyline and carbamazepine in the treatment
of dysesthesia pain in spinal cord injury. Arch Phys
Medic Rehab 73, 300
301.
Sang CN, Miller VA, Dobosch L, Gracely RH,
Hayden D (1999) Temporal summation and
reduction of pain thresholds in normal skin in

References
patients with dysesthetic central pain following
spinal cord injury. In 9th World Congress on Pain,
Book of Abstracts. Seattle, WA: IASP Press, A114.
Sang CN (2002) Glutamate receptor antagonists in
central neuropathic pain following spinal cord
injury. Progr Pain Res Manage 23, 365
377.
Sano K, Yoshioka M, Ogashiwa M, Ishijima B,
Ohye C (1966) Thalamolaminotomy. A new
operation for relief of intractable pain. Conf
Neurol 27, 63
66.
Sano K (1977) Intralaminar thalamotomy (thalamolaminotomy), and posteromedial hypothalamotomy in the treatment of intractable pain. In
Krayenbuhl H, Maspes PE, Sweet WH, eds.,
Progress in Neurological Surgery. Pain
Its
Neurosurgical Management: II. Central Procedures.
Basel: Karger, pp. 50
103.
Sasaki K (1938) Ueber die Wirkung der Chordotomie auf Spontangangraen. Arch Klin Chir 192,
448
461.
Savas A, Kanpolat Y (2005) Comment to Brown JA
and Pilitsis JG. Neurosurgery 56, 296.
Scarff JE (1950) Unilateral prefrontal lobotomy for
the relief of intractable pain. Report of 58 cases
with special consideration of failures. J Neurosurg
7, 330
336.
Schludermann E, Zubek JP (1962) Effect of age on
pain sensitivity. Percept Skills 14, 295
301.
Schmahmann JD, Leifer D (1992) Parietal pseudothalamic pain syndrome. Clinical features and
anatomic correlates. Arch Neurol 49, 1032
1037.
Schmahmann JD (2003) Vascular syndromes of the
thalamus. Stroke 34, 2264
2278.
Schnitzler A, Ploner M (2000) Neurophysiology and
functional neuroanatomy of pain perception.
J Clin Neurophysiol 17, 592
603.
Schoenen J, Grant G (2004) Spinal cord: connections. In Paxinos G, Mai JK, eds., The Human
Nervous System, 2nd edn. Amsterdam: Elsevier
Academic Press, pp. 233
250.
Scholz J, Vieregge P, Moser A (1999) Central pain
as a manifestation of partial epileptic seizures.
Pain 80, 445
450.
Schott B, Laurent B, Mauguire F (1986) Les douleurs
thalamiques. E`tude critique de 43 cas. Rev Neurol
142, 308
315.
Schott GD, Loh L (1984) Anticholinesterase drugs in
the treatment of chronic pain. Pain 20, 201
206.
Schott GD (1996) From thalamic syndrome to
central poststroke pain. J Neurol Neurosurg Psych
61, 560
564.
Schott GD (2001a) Delayed onset and resolution of
pain. Some observations and implications. Brain
124, 1067
1076.

369
Schott GD (2001b) Nosological entities? Reflex
sympathetic dystrophy. J Neurol Neurosurg Psych
71, 291
295.
Schvarcz JR (1977) Periaqueductal mesencephalotomy for facial central pain. In Sweet
WH, Obrador S, Martin-Rodriguez JG, eds.,
Neurosurgical Treatment in Psychiatry, Pain and
Epilepsy. Baltimore, MD: University Park Press,
pp. 661
667.
Schvarcz JR (1978) Spinal cord stereotactic techniques for trigeminal nucleotomy and extralemniscal myelotomy. Appl Neurophysiol 41, 99
112.
Schvarcz JR (1980) Chronic self-stimulation of the
medial posterior inferior thalamus for the alleviation of deafferentation pain. Acta Neurochir
Suppl 20, 295
301.
Schwartz HG (1950) Neurosurgical relief of intractable pain. Surg Clin N Amer 30, 1379
1389.
Schwartz HG (1960) High cervical tractotomy:
technique and results. Clin Neurosurg 8,
282
293.
Schwoebel J, Coslett HB, Bradt J, Friedman R,
Dileo C (2002) Pain and the body schema: effects
of pain severity on mental representations of
movement. Neurology 59, 775
777.
Sedan R, Lazorthes Y (1978) La neurostimulation
e`lectrique the`rapeutique. Soc. Neurochir. Langue
Franc. XXVIII Congres Annuel, Neurochirurgie,
pp. 1
138.
Seghier ML, Lazayras F, Vuilleumier P, Schnider A,
Carota A (2005) Functional magnetic resonance
imaging and diffusion tensor imaging in a case
of central poststroke pain. J Pain 6, 208
212.
Selverstone AI, Moulins M (1985) Oscillatory neural
networks. Annu Rev Physiol 47, 29
48.
Semyanov A, Walker MC, Kullmann DM, Silver RA
(2004) Tonically active GABA A receptors:
modulating gain and maintaining the tone. TINS
27, 262
269.
Serpell MG, for the Neuropathic Pain Study Group
(2002) Gabapentin in neuropathic pain syndromes: a randomised, double-blind, placebocontrolled trial. Pain 99, 557
566.
Shapiro PE, Braun CW (1987) Unilateral pruritus
after a stroke. Arch Dermatol 123, 1527
1530.
Shepherd GM (2004) The Synaptic Organization of
the Brain. Oxford: Oxford University Press.
Shergill SS, Bays PM, Frith CD, Wolpert DM (2003)
Two eyes for an eye: the neuroscience of force
escalation. Science 301, 187.
Sherman SM, Guillery RW (2004) Thalamus. In
Shepherd GM, ed., The Synaptic Organization of
the Brain. Oxford: Oxford University Press,
pp. 311
360.

370
Shibasaki H, Kuroiwa Y (1974) Painful tonic seizures
in multiple sclerosis. Arch Neurol 30, 47
51.
Shieff C, Nashold BS (1988) Thalamic pain and
stereotactic mesencephalotomy. Acta Neurochir
(Wien) Suppl 42, 239
242 (same series appearing
in Br J Neurosurg 1987, 1, 305
310 and Neurol Res
1987, 9, 101
104).
Shieff C (1991) Treatment of central deafferentation
syndromes: thalamic syndrome. In Nashold BS Jr.,
Ovelmen-Levitt J, eds., Deafferentation Pain
Syndromes. Pathophysiology and Treatment,
New York: Raven Press, pp. 285
290.
Shimodozono M, Kawahira K, Kamishita T et al.
(2002) Reduction of central poststroke pain with
the selective serotonin reuptake inhibitor fluvoxamine. Int J Neurosci 112, 1173
1181.
Shulman R, Turnbull IM, Diewold P (1982)
Psychiatric aspects of thalamic stimulation for
neuropathic pain. Pain 13, 127
135.
Shulman RG, Rothman DL, Behar KL, Hyder F
(2004) Energetic basis of brain activity: implications for neuroimaging. TINS 27, 489
495.
Sicuteri F (1971) Reversible central pain syndrome
in man following treatment with P-chlorophenylalanine and reserpine. Pharmacol Res Commun
3, 401.
Siddall PJ, Gray M, Rutkowski S, Cousins MJ (1994)
Intrathecal morphine and clonidine in the management of spinal cord injury pain: a case report.
Pain 59, 147
148.
Siddall PJ, Taylor DA, McClelland JM, Rutkowski SB,
Cousins MJ (1999) Pain report and the relationship of pain to physical factors in the first 6 months
following spinal cord injury. Pain 81, 187
197.
Siddall PJ, Molloy AR, Walker S et al. (2000) The
efficacy of intrathecal morphine and clonidine in
the treatment of pain after spinal cord injury.
Anesth Analg 9, 1493
1498.
Siddall PJ (2002) Spinal drug administration in the
treatment of spinal cord injury pain. Progr Pain
Res Manage 23, 353
364.
Siegfried J, Krayenbuhl H (1972) Clinical experience
in the treatment of intractable pain. In Janzen R,
Keidel WD, Herz A, Steichele C, eds., Pain: Basic
Principles
Pharamacology
Therapy. Stuttgart:
Georg Thieme, pp. 202
204.
Siegfried J (1977) Stereotactic pulvinarotomy in the
treatment of intractable pain. Progr Neurol Surg
8, 101
113.
Siegfried J (1991) Therapeutical neurostimulation:
indications reconsidered. Acta Neurochir (Wien)
Suppl 52, 112
117.
Sillito AM, Jones HE, Gerstein GL, West DC (1994)
Feature-linked synchronization of thalamic relay

References
cell firing induced by feedback from the visual
cortex. Nature 369, 479
482.
Silver ML (1957) Central pain from cerebral
arteriovenous aneurysm. J Neurosurg 14, 92
96.
Silverman IE (1998) Central poststroke pain associated with lateral medullary infarction. Neurology
50, 836
837.
Simpson BA (1991) Spinal cord stimulation in
60 cases of intractable pain. J Neurol Neurosurg
Psych 54, 196
199.
Simpson BA (1999) Spinal cord stimulation. Pain
Rev 1, 199
230.
Sindou M, Keravel Y (1980) Analgesie par la
methode delectrostimulation transcutanee. Resultats dans les douleurs dorigine neurologique.
A propos de 180 cas. Neurochirurgie 26, 153
157.
Sindou MP, Mertens P, Garcia-Larrea L (2001)
Surgical procedures for neuropathic pain. Neurosurg Quart 11, 45
65.
Sindou M, Mertens P, Bendavid U, Garcia-Larrea L,
Mauguiere F (2003) Predictive value of somatosensory evoked potentials for long-lasting pain
relief after spinal cord stimulation: practical use
for patient selection. Neurosurgery 52, 1374
1384.
Sindrup SH, Jensen TS (1999) Efficacy of pharmacological treatments of neuropathic pain: an
update and effect related to mechanism of drug
action. Pain 83, 389
400.
Sist TC, Filadora VA II, Miner M, Lema M (1997)
Experience with gabapentin for neuropathic pain
in the head and neck: report of ten cases. Reg
Anesth 22, 473
478.
Sjolund BH (1991) Role of transcutaneous electrical
nerve stimulation, central nervous system stimulation, and ablative procedures in central pain
syndromes. In Casey KL, ed., Pain and Central
Nervous System Disease. The Central Pain
Syndromes. New York: Raven Press, pp. 267
275.
Sjo lund BH (1993) Transcutaneous electrical
stimulation in neuropathic pain. Pain Digest 3,
23
26.
Sjoqvist O (1950) La section chirurgicale des cordons
et des voies de la douleur dans la moelle et le tronc
cerebral. Rev Neurol 83, 38
40.
Slaughter RF (1938) Relief of causalgic-like pain in
the isolated extremity by sympathectomy. JMA
Georgia 27, 253
256.
Slawek J, Reclowicz D, Zielinski P, Sloniewski P,
Nguyen JP (2005) [Motor cortex stimulation in
the central pain syndrome]. Neurol Neurochir Pol
39, 237
240.
Sleghart W, Sperk G (2002) Subunit composition,
distribution and function of GABA A receptor
subunits. Curr Top Med Chem 2, 795
816.

References
Slonimski M, Abram SE, Zuniga RE (2004) Intrathecal baclofen in pain management. Reg
Anesth Pain Med 29, 269
276.
Smith AJ, Blumenfeld H, Behar KL et al. (2002)
Cerebral energetics and spiking frequency: the
neurophysiological basis of fMRI. Proc Natl Acad
Sci USA 99, 10765
10770.
Smith KJ, McDonald WI (1982) Spontaneous and
evoked electrical discharges from a central demyelinating lesion. J Neurol Sci 55, 39
47.
Smolik FA, Nash FP, Machek O (1960) Spinal
cordectomy in the management of spastic paraplegia. Am Surg 26, 639
645.
Solaro C, Brichetto G, Amato MP et al., and the
PaIMS Study group (2004) The prevalence of pain
in multiple sclerosis. A multicenter cross-sectional
study. Neurology 63, 919
921.
Soria ED, Fine EJ (1991) Disappearance of thalamic
pain after parietal subcortical stroke. Pain 44,
285
288.
Soros P, Imai T, Bantel C et al. (1999) Plasticity of
the somatosensory cortex during cluster headache
attacks. In 9th World Congress on Pain, Book of
Abstracts. Seattle, WA: IASP Press, Abst. 91-P403.
Sourek K (1969) Commissural myelotomy. J Neurosurg 31, 524
527.
Spaic M, Markovic N, Tadic R (2002) Microsurgical
DREZotomy for pain of spinal cord and cauda
equina injury origin: clinical characteristics of
pain and implications for surgery in a series
of 26 patients. Acta Neurochir (Wien) 144,
453
462.
Spiegel EA, Wycis HT, Freed H (1952) Stereoencephalotomy. Thalamotomy and related procedures.
JAMA 148, 446
451.
Spiegel EA, Kletzkin M, Szekely EG, Wycis HT
(1954) Role of hypothalamic mechanisms in
thalamic pain. Neurology 4, 735
751.
Spiegel EA, Wycis HT (1962) Stereoencephalotomy:
II. Clinical and Physiological Applications.
New York: Grune & Stratton.
Spiegel EA, Wycis HT, Szekely EG, Gildenberg PL
(1966) Medial and basal thalamotomy in so-called
intractable pain. In Knighton RS, Dumke PR, eds.,
Pain. Boston, MA: Little Brown, pp. 503
517.
Spiegelmann R, Friedman WA (1991) Spinal cord
stimulation: a contemporary series. Neurosurgery
28, 65
71.
Stein BE, Price DD, Gazzaniga MS (1989) Pain
perception in a man with total corpus callosum
transection. Pain 38, 51
56.
Steinke W, Sacco RL, Mohr JP et al. (1992) Thalamic
stroke. Presentation and prognosis of infarcts and
hemorrhages. Arch Neurol 49, 703
710.

371
Stenager E, Knudsen L, Jensen K (1991) Acute and
chronic pain syndromes in multiple sclerosis. Acta
Neurol Scand 84, 197
200.
Steriade M, Jones EG, McCormick DA (1997)
Thalamus. Amsterdam: Elsevier.
Steriade M (1999) Coherent oscillations and shortterm plasticity in corticothalamic networks. TINS
22, 337
345.
Stewart I (1997) Does God Play Dice? The New
Mathematics of Chaos. London: Penguin Books.
Stewart WA, Stoops WL, Pillone PR, King RB (1964)
An electrophysiologic study of ascending pathways
from nucleus caudalis of the spinal trigeminal
nuclear complex. J Neurosurg 21, 35
48.
Stoermer S, Gerner HJ, Grueninger W et al. (1997)
Chronic pain/dysaesthesiae in spinal cord injury
patients: results of a multicentre study. Spinal
Cord 35, 446
455.
Stone TT (1950) Phantom limb pain and central
pain. Relief by ablation of portion of posterior
central cerebral convolution. Arch Neurol Psych
63, 739
748.
Stoodley MA, Warren JD, Oatey PE (1995) Thalamic
syndrome caused by unruptured cerebral aneurysm. Case report. J Neurosurg 82, 291
293.
Strogatz S (2003) Sync: The Emerging Science of
Spontaneous Order. New York: Hyperion.
Sugita K, Mutsuga N, Rakaoka T, Doi T (1972)
Results of stereotaxic thalamotomy for pain.
Conf Neurol 34, 265
274.
Sullivan MJ, Drake ME Jr. (1984) Unilateral pruritus
and nocardia brain abscess. Neurology 34,
828
829.
Supe`r H, Spekreijse H, Lamme VAF (2001) Two
distinct modes of sensory processing observed in
monkey primary visual cortex (V1). Nature
Neurosci 4, 304
310.
Suzuki M, Davis C, Symon L, Gentili F (1985)
Syringoperitoneal shunt for treatment of
cord cavitation. J Neurol Neurosurg Psych 48,
620
627.
Svendsen KB, Jensen TS, Bach FW (2004) Does the
cannabinoid dronabinol reduce central pain in
multiple sclerosis? Randomised double blind
placebo controlled crossover trial. Br Med J
329, 253.
Svendsen KB, Jensen TS, Hansen HJ, Bach FW
(2005) Sensory function and quality of life in
patients with multiple sclerosis and pain. Pain
114, 473
481.
Svensson P, Minoshima S, Beydoun A, Morrow TJ,
Casey KL (1997) Cerebral processing of acute skin
and muscle pain in humans. J Neurophysiol 78,
450
460.

372
Sweet WH, Wepsic JG (1974) Stimulation of pain
suppressor mechanisms. A critique of some
current methods. Adv Neurol 4, 734
746.
Sweet WH, Wepsic J (1975) Stimulation of the
posterior columns of the spinal cord for pain
control. Surg Neurol 4, 133.
Sweet WH, Poletti CE (1989) Operations in the
brain stem and cord, with an appendix on open
cordotomy. In Wall PD, Melzack R, eds., Textbook
of Pain, 2nd edn. Edinburgh: Churchill Livingstone, pp. 811
831.
Sweet WH (1991) Deafferentation syndromes in
humans: a general discussion. In Nashold BS Jr.,
Ovelmen-Levitt J, eds., Deafferentation Pain
Syndromes. Pathophysiology and Treatment,
New York: Raven Press, pp. 259
274.
Szentirmai O, Carter BS (2004) Genetic and cellular
therapies for cerebral infarction. Neurosurgery 55,
283
297.
Tai Q, Kirschblum S, Chen B et al. (2002)
Gabapentin in the treatment of neuropathic pain
after spinal cord injury: a prospective, randomized, double-blind, crossover trial. J Spinal Cord
Med 25, 100
105.
Taira T, Tanikawa T, Kawamura H, Iseki H,
Takakura K (1994) Spinal intrathecal baclofen
suppresses central pain after a stroke. J Neurol
Neurosurg Psych 57, 381
382.
Taira T, Kawamura H, Tanikawa T et al. (1995) A
new approach to control central deafferentation
pain: spinal intrathecal baclofen. Stereotact Funct
Neurosurg 65, 101
105.
Taira T (1998) Comments on Eisenberg and Pud,
Pain 74, 337
339. Pain 78, 221
226.
Takahashi Y, Hashimoto K, Tsuji S (2004) Successful
use of zonisamide for central poststroke pain.
J Pain 5, 192
194.
Takano M, Takano Y, Sato I (1999) The effect of oral
amantadine in chronic pain patients with positive
ketamine challenge test. In 9th World Congress on
Pain, Book of Abstracts. Seattle, WA: IASP Press,
Abst. 220-P68.
Talbot JD, Villemure JG, Bushnell MC, Duncan GH
(1995) Evaluation of pain perception after anterior capsulotomy: a case report. Somatosens Mot
Res 12, 115
126.
Talairach J, Hecaen H, David M, Monnier M,
De Ajuraguerra J (1949) Recherches sur la
coagulation the`rapeutique des structures souscorticales chez lhomme. Rev Neurol 81, 4
24.
Talairach J (1955) Chirurgie stereotaxique du
thalamus. Bases anatomiques et techniques. Indications et resultats therapeutiques. VI Congr Lat
Am Neurocir (Montevideo), pp. 865
925.

References
Talairach J, Tournoux P, Bancaud J (1960) La
chirurgie parietale de le douleur. Acta Neurochir
(Wien) 8, 153
250.
Tasker RR, Organ LW, Hawrylyshyn P (1980)
Deafferentation and causalgia. In Bonica JJ, ed.,
Pain. New York: Raven Press, pp. 305
330.
Tasker RR, Tsuda T, Hawrylyshyn P (1983) Clinical
neurophysiological investigation of deafferentation pain. Adv Pain Res Ther 5, 713
738.
Tasker RR (1984) Deafferentiation. In Wall PD,
Melzack R, eds., Textbook of Pain. London:
Churchill Livingstone, pp. 119
132.
Tasker RR, Yoshida M, Sima AAF, Deck J (1986)
Stimulation mapping of the periventricularperiaqueductal gray (PVG-PAG) in man: an
autopsy study. In Samii M, ed., Surgery In and
Around the Brain Stem and the Third Ventricle.
Berlin: Springer Verlag, pp. 161
167.
Tasker RR, Dostrovsky J (1989) Deafferentation
and central pain. In: Wall PD, Melzack R, eds.,
Textbook of Pain. London: Churchill Livingstone,
pp. 154
180.
Tasker RR (1989) Stereotactic surgery. In Wall PD,
Melzack R, eds., Textbook of Pain, 2nd edn.
Edinburgh: Churchill Livingstone, pp. 840
855.
Tasker RR (1990) Thalamotomy. Neurosurg Clin N
Am 1, 84
86.
Tasker RR, DeCarvalho G, Dostrovsky JO (1991)
The history of central pain syndromes, with observations concerning pathophysiology and treatment. In Casey KL, ed., Pain and Central Nervous
System Disease: The Central Pain Syndromes.
New York: Raven Press, pp. 31
58.
Tasker RR, DeCarvalho GTC, Dolan EJ (1992)
Intractable pain of spinal cord origin: clinical
features and implications for surgery. J Neurosurg
77, 373
378.
Tasker RR, Parrent AG, Kiss Z, Davis K, Dostrovsky
JO (1994) Surgical treatment of stroke-induced
pain. Stereotact Funct Neurosurg 62, 311
312.
Tasker RR, North RB (1997) Cordotomy and
myelotomy. In North RB, Levy RM, eds., Neurosurgical Management of Pain. New York: Springer,
pp. 191
220.
Tasker RR (2001a) Central pain states. In Loeser JD,
ed., Bonicas Management of Pain, 3rd edn.
Philadelphia, PA: Lippincott Williams & Wilkins,
pp. 433
457 (see also previous edn., 1990).
Tasker RR (2001b) Microelectrode findings in the
thalamus in chronic pain and other conditions.
Stereotact Funct Neurosurg 77, 166
168.
Tator CH, Agbi CB (1991) Complications in the
management of syringomyelia. Perspect Neurol
Surg 2, 143
150.

References
Taub E, Munz M, Tasker RR (1997) Chronic
electrical stimulation of the gasserian ganglion
for the relief of pain in a series of 34 patients.
J Neurosurg 86, 197
202.
Teixeira MJ (1998) Various functional procedures
for pain: II. Facial pain. In Gildenberg PL,
Tasker RR, eds., Textbook of Stereotactic and
Functional Neurosurgery. New York: McGraw-Hill,
pp. 1389
1402.
Teixeira MJ, Lepski G, Aguiar PHP et al. (2003)
Bulbar trigeminal stereotactic nucleotractotomy
for treatment of facial pain. Stereotact Funct
Neurosurg 81, 37
42.
Thalhammer JG, Raymond SA (1991) Does the axon
only ax? IASP newsletter, July
August, pp. 2
4.
Thomas DGT, Jones SJ (1984) Dorsal root entry
zone lesions (Nasholds procedure) in brachial
plexus avulsion. Neurosurgery 15, 966
968.
Thompson SC (1981) Will it hurt less if I can control
it? A complex answer to a simple question. Psychol
Bull 90, 89
101.
Thomson AM, Deuchars J (1994) Temporal and
spatial properties of local circuits in neocortex.
TINS 17, 119
126.
Tirsch WS, Stude P, Scherb H, Keidel M (2004)
Temporal order of nonlinear dynamics in human
brain. Brain Res Rev 45, 79
95.
Tommerdahl M, Delemos KA, Vierck CJ Jr., Favorov
OV, Whitsel BL (1996) Anterior parietal cortical
response to tactile and skin-heating stimuli
applied to the same skin site. J Neurophysiol 75,
2662
2670.
Tommerdahl M, Delemos KA, Favorov OV et al.
(1998) Response of anterior parietal cortex to
different modes of same-site skin stimulation.
J Neurophysiol 80, 3272
3281.
Tommerdahl M, Whitsel BL, Favorov OV, Metz
CB, OQuinn BL (1999) Responses of contralateral SI and SII in cat to same-site cutaneous
flutter versus vibration. J Neurophysiol 82,
1982
1992.
Tommerdahl M, Delemos KA, Whitsel BL, Favorov
OV, Metz CB (1999) Response of anterior
parietal cortex to cutaneous flutter versus vibration. J Neurophysiol 82, 16
33.
Torvik A (1959) Sensory, motor and reflex changes
in two cases of intractable pain after stereotactic
mesencephalic tractotomy. J Neurol Neurosurg
Psychiat 22, 299
305.
Toth S, Solyom A, Toth Z (1984) One possible
mechanism of central pain. Autokindling phenomenon on the phantom limb or sensory loss
oriented patients. Acta Neurochir (Wien) Suppl 33,
459
469.

373
Tourian AY (1987) Narcotic responsive thalamic
pain treatment with propranolol and tricyclic
antidepressants. Pain Suppl 4, 411.
Tourian AY (1991) Deafferentation syndrome:
medical treatment. In: Nashold BS Jr.,
Ovelmen-Levitt J, eds., Deafferentation Pain Syndromes. Pathophysiology and Treatment, New
York: Raven Press, pp. 331
340.
Tovi D, Schisano G, Liljequist B (1961) Primary
tumors of the region of the thalamus. J Neurosurg
18, 730
740.
Treede R-D, Bromm B (1991) Neurophysiological
approaches to the study of spinothalamic tract
function in humans. In Casey KL, ed., Pain and
Central Nervous System Disease. New York: Raven
Press, pp. 117
127.
Tremont-Lukats IW, Challapalli V, McNicol ED,
Lau J, Carr DB (2005) Systemic administration of
local anesthetics to relieve neuropathic pain:
a systematic review and meta-analysis. Anesth
Analg 101, 1738
1749.
Trentin L, Visentin M (2000) La predittivita` del test
con lidocaina nel trattamento del dolore neuropatico. Minerva Anestesiol 66, 157
161.
Triggs WJ, Beric A (1992) Sensory abnormalities
and dysaesthesias in the anterior spinal artery
syndrome. Brain 115, 189
198.
Triggs WJ, Beric A (1993) Giant somatosensory
evoked potentials in a patient with the anterior
spinal artery syndrome. Muscle Nerve 16, 492
497.
Triggs WJ, Beric A (1994) Dysaesthesiae induced by
physiological and electrical activation of posterior
column afferents after stroke. J Neurol Neurosurg
Psych 57, 1077
1080.
Tsai PS, Buerkle H, Huang LT et al. (1998) Lidocaine
concentrations in plasma and cerebrospinal fluid
after systemic bolus administrations in humans.
Anesth Analg 87, 601
604.
Tseng SH (2000) Treatment of chronic pain by
spinal cord stimulation. J Formos Med Assoc 99,
267
271.
Tsubokawa T, Moriyasu N (1975) Follow-up results
of centre median thalamotomy for relief of
intractable pain. Conf Neurol 37, 280
284.
Tsubokawa T, Yamamoto T, Katayama T, Hirayama
T, Sibuya H (1984) Thalamic relay nucleus
stimulation for relief of intractable pain. Clinical
results and beta-endorphin immunoreactivity in
the cerebrospinal fluid. Pain 18, 115
126.
Tsubokawa T, Katayama Y, Yamamoto T, Hirayama
T (1985) Deafferentation pain and stimulation
of the thalamic sensory relay nucleus: clinical and
experimental study. Appl Neurophysiol 48,
166
171.

374
Tsubokawa T, Katayama Y, Yamamoto T, Hirayama
T, Koyama S (1991) Treatment of thalamic pain
by chronic motor cortex stimulation. PACE 14,
131
134.
Tsubokawa T, Katayama Y, Yamamoto T, Hirayama
T, Koyama S (1993) Chronic motor cortex
stimulation in patients with thalamic pain.
J Neurosurg 78, 393
401.
Turnbull F (1939) Cordotomy for thalamic pain.
A case report. Yale J Biol Med 2, 411
414.
Turnbull IM (1972) Bilateral cingulumotomy combined with thalamotomy or mesencephalic
tractotomy for pain. Surg Gynecol Obstet 134,
958
962.
Turnbull IM (1984) Brain stimulation. In Wall PD,
Melzack R, eds., Textbook of Pain, Edinburgh:
Churchill Livingstone, pp. 706
714.
Turner JA, Lee JS, Schandler SL, Cohen MJ (2003)
An fMRI investigation of hand representation in
paraplegic humans. Neurorehab Neural Repair
17, 37
47.
Ullman S (1995) Sequence seeking and counter
streams: a computational model for bidirectional
information flow in the visual cortex. Cereb Cortex
5, 1
11.
Urabe M, Tsubokawa T (1965) Stereotaxic thalamotomy for the relief of intractable pain. Tohoku
J Exp Med 85, 286
300.
Urban BJ, Nashold BS Jr. (1978) Percutaneous
epidural stimulation of the spinal cord for relief
of pain. J Neurosurg 48, 323
328.
Usubiaga J, Moya F, Wikinski J (1967) Relationship
between the passage of local anesthetics across
the blood brain barrier and their effects on the
central nervous system. Br J Anaesth 39, 943
947.
Van Bastelaere M, De Laat M (1999) Lamotrigine:
a morphine-sparing drug for central pain. In 9th
World Congress on Pain, Book of Abstracts. Seattle,
WA: IASP Press, A211-P65.
Van der Bruggen MA, Huisman HB, Beckermann H
et al. (2001) Randomized trial of 4-aminopyridine
in patients with chronic incomplete spinal cord
injury. J Neurol 248, 665
671.
Vega-Bermudez F, Johnson KO (2002) Spatial acuity
after digit amputation. Brain 125, 1256
1264.
Velasco M, Brito F, Jimenez F et al. (1998) Effect of
fentanyl and naloxone on a thalamic induced
painful response in intractable epileptic patients.
Stereotact Funct Neurosurg 71, 90
102.
Veldman PHJM, Goris RJ (1996) Multiple reflex
sympathetic dystrophy. Which patients are at risk
for developing a recurrence of reflex sympathetic
dystrophy in the same or another limb? Pain 64,
463
466.

References
Verdugo R, Ochoa JL (1991) High incidence of
placebo responders among chronic neuropathic
pain patients. Ann Neurol 30, 229.
Verhoeff NP, Petroff OA, Hyder F et al. (1999)
Effects of vigabatrin on the GABAergic system
as determined by [123I] iomazenil SPECT and
GABA MRS. Epilepsia 40, 1433
1438.
Vestergaard K, Nielsen J, Andersen G et al. (1995)
Sensory abnormalities in consecutive, unselected
patients with central post-stroke pain. Pain 61,
177
186.
Vestergaard K, Andersen G, Jensen TS (1996)
Treatment of central post-stroke pain with a
selective serotonin reuptake inhibitor. Eur J Neurol
3(Suppl. 5), 169.
Vestergaard K, Andersen G, Gottrup H, Kristensen
BT, Jensen TS (2000) Lamotrigine for central
poststroke pain: a randomized controlled trial.
Neurology 56, 184
190.
Vick PG, Lamer TJ (2001) Treatment of central poststroke pain with oral ketamine. Pain 92, 311
313.
Vierck CJ (1973) Alterations of spatio-tactile discrimination after lesions of primate spinal cord.
Brain Res 58, 69
79.
Vilela Filho O, Tasker RR (1994) Pathways involved
in thalamic ventrobasal stimulation for pain relief:
evidence against the hypothesis VB stimulationrostroventral medulla excitation-dorsal horn
inhibition. Arq Neuropsiquiatr 52, 386
391.
Vilela Filho O (1996) Risk factors for unpleasant
paresthesiae induced by paresthesiae-producing
deep brain stimulation. Arq Neuropsiquiatr 54,
57
63.
Villemure C, Wassimi S, Bennett GJ, Shir Y, Bushnell
MC (2006) Unpleasant odors increase pain
processing in a patient with neuropathic pain:
psychophysical and fMRI investigation. Pain 120,
213
220.
Vogel HP, Heppner B, Humbs N, Schramm J,
Wagner C (1986) Long term effects of spinal cord
stimulation in chronic pain syndromes. J Neurol
233, 16
18.
Vogt BA, Berger GR, Derbyshire SW (2003) Structural and functional dichotomy of human midcingulate cortex. Eur J Neurosci 18, 3134
3144.
Von Hagen KO (1957) Chronic intolerable pain.
Discussion of its mechanism and report of
eight cases treated with electroshock. JAMA 165,
773
777.
Voris HC, Whisler WW (1975) Results of stereotaxic
surgery for intractable pain. Conf Neurol 37,
86
96.
Vranken JH, Dijkgraaf MG, Kruis MR, van Dasselaar
NT, van der Vegt MH (2005) Iontophoretic

References
administration of S()-ketamine in patients with
intractable central pain: a placebo-controlled trial.
Pain 118, 224
231.
Vuadens PH, Regli F, Dolivo M, Uske A (1994)
Segmental pruritus and intramedullary vascular
formation. Schw Arch Neurol Psych 145, 13
16.
Wade DT, Makela P, Robson P, House H, Bateman
C (2004) Do cannabis-based medicinal extracts
have general or specific effects on symptoms in
multiple sclerosis? A double-blind, randomized,
placebo-controlled study on 160 patients. Mult
Scler 10, 434
441.
Waijima Z, Shitara T, Inoue T, Ogawa R (2000)
Severe lightning pain after subarachnoid block in a
patient with neuropathic pain of central origin:
which drug is best to treat the pain? Clin J Pain 16,
265
269.
Walker AE (1942a) Relief of pain by mesencephalic
tractotomy. Arch Neurol Psych 48, 865
883.
Walker AE (1942b) Mesencephalic tractotomy.
A method for the relief of unilateral intractable
pain. Arch Surg 44, 953
962.
Walker AE (1950) The neurosurgical treatment
of intractable pain. Lancet 70, 279
282.
Walker AE (1955) Pain: the neurosurgeons viewpoint. J Chron Dis 2, 91
95.
Wall PD (1995) Pain in the brain and lower parts
of the anatomy. Pain 62, 389
391.
Walshe TM, Davis KR, Fisher CM (1977) Thalamic
hemorrhage: a computed tomographic
clinical
correlation. Neurology 27, 217
222.
Waltz TA, Ehni G (1966) The thalamic syndrome
and its mechanisms. Report of two cases, one due
to arteriovenous malformation in the thalamus.
J Neurosurg 24, 735
742.
Wang XJ (2001) Synaptic reverberation underlying
mnemonic persistent activity. Trends Neurosci 24,
455
463.
Warms CA, Turner JA, Marshall HM, Cardenas DD
(2002) Treatments for chronic pain associated
with spinal cord injuries: many are tried, few are
helpful. Clin J Pain 18, 154
163.
Waxman SG, Hains BC (2006) Fire and phantoms
after spinal cord injury: Na() channels and
central pain. Trends Neurosci 29, 207
215.
Weigel R, Krauss JK (2004) Center medianparafascicular complex and pain control. Stereotact Funct Neurosurg 82, 115
126.
Weiller C, Chollet F, Friston KJ, Wise RJ, Frackowiak
RS (1992) Functional reorganization of the
brain in recovery from striatocapsular infarction
in man. Ann Neurol 3, 463
472.
Weimar C, Kloke M, Schlott M, Katsarava Z,
Diener HC (2002) Central poststroke pain in

375
a consecutive cohort of stroke patients. Cerebrovasc Dis 14, 261
263.
Weisenberg M, Schwarzwald J, Tepper I (1996) The
influence of warning signal timing and cognitive
preparation on the aversiveness of cold-pressor
pain. Pain 64, 379
385.
Werhahn KJ, Mortensen J, Kaelin-Lang A,
Boroojerdi B, Cohen LG (2002) Cortical
excitability changes induced by deafferentation
of the contralateral hemisphere. Brain 125,
1402
1413.
Werner A (1961) Myelectomie dans un cas de
paraplegie spastique douloureux. Neurochirurgie
7, 140
145.
Wertheimer P, Mansuy L (1949) Reflexions sur la
topectomie prefrontale. Rev Neurol 81, 866
871.
(see also 401
408)
Wertheimer P, Lecuire J (1953) La myelotomie
commissurale posterieure: a` propos de 107
observations. Acta Chir Belg 52, 568
574.
Wessel K, Vieregge P, Kessler CH, Kompf D (1994)
Thalamic stroke: correlation of clinical symptoms,
somatosensory evoked potentials, and CT findings. Acta Neurol Scand 90, 167
173.
Wester K (1987) Dorsal column stimulation in pain
treatment. Acta Neurol Scand 75, 151
155.
White JC, Sweet WH (1955) Pain: Its Mechanisms
and Neurosurgical Control. Springfield, IL:
Thomas.
White JC (1962) Modifications of frontoleukotomy for relief of pain and suffering in
terminal malignant diseases. Ann Surg 156,
394
403.
White JC (1963) Anterolateral chordotomy
its
effectiveness in relieving pain of non-malignant
disease. Neurochirurgia (Stuttgart) 6, 83
102.
White JC, Sweet WH (1969) Pain and the Neurosurgeon. A Forty-Year Experience. Springfield, IL:
Thomas.
Widar M, Samuelsson L, Karlsson-Tivenius S,
Ahlstroem G (2002) Long-term pain conditions
after a stroke. J Rehabil Med 34, 165
170.
Widar M, Ek A-C, Ahlstroem G (2004) Coping with
long-term pain after a stroke. J Pain Symptom
Manage 27, 215
225 (see also J Clin Nurs 2004,
13, 497
505).
Widerstroem-Noga E, Felipe-Cuervo E, Yezierski RP
(2001) Chronic pain after spinal injury: interference with sleep and daily activities. Arch Phys
Med Rehabil 82, 1571
1577.
Wiegand H, Winkelmuller W (1985) Behandlung des
Deafferentierungsschmerzes durch Hochfrequenzlasion der Hinterwurzeleintrittszone. Deutsche
medizinische Wochenschrift 110, 216
220.

376
Wiffen P, Collins S, McQuay H et al. (2000)
Anticonvulsant drugs for acute and chronic pain.
Cochrane Database Syst Rev CD001133.
Willis WD (1985) The Pain System. The Neural Basis
of Nociceptive Transmission in the Mammalian
Nervous System. Basel: Karger.
Willis WD (1991) Central neurogenic pain: possible
mechanisms. Adv Pain Res Ther 19, 81
102.
Willis WD (2002) Possible mechanisms of central
neuropathic pain. Progr Pain Res Manage 23,
85
115.
Willis WD, Westlund KN (2004) Pain system. In
Paxinos G, Mai JK, eds., The Human Nervous
System, 2nd edn. Amsterdam: Elsevier Academic
Press, pp. 1137
1170.
Willoch F, Schindler F, Wester HJ et al. (2004)
Central poststroke pain and reduced opioid
receptor binding within pain processing circuitries: a [11C] diprenorphine PET study. Pain 108,
213
220.
Winkelmuller M, Winkelmuller W (1996) Longterm effects of continuous intrathecal opioid
treatment in chronic pain of nonmalignant
etiology. J Neurosurg 85, 458
467.
Wirth ED III, Vierck CJ Jr., Reier PJ, Fessler RG,
Anderson DK (2002) Correlation of MRI findings
with spinal cord injury pain following neural
tissue grafting into patients with posttraumatic
syringomyelia. Progr Pain Res Manage 23,
313
330.
Witting N, Kupers RC, Svensson P et al. (2001)
Experimental brush-evoked allodynia activates
posterior parietal cortex. Neurology 57,
1817
1824.
Wolskee PJ, Gracely RH, Greenberg RP, Dubner R,
Lees D (1982) Comparison of effects of morphine
and deep brain stimulation on chronic pain.
Am Pain Soc Abst 36.
Wood T, Sloan R (1997) Successful use of ketamine
for central pain. Palliat Med 11, 57.
Woodrow KM, Friedman GD, Siegelaub AB, Collen
MF (1972) Pain tolerance: differences according
to age, sex and race. Psychsom Med 34, 548
556.
Woolsey CN, Erickson TC, Gilson WE (1979)
Localization in somatic sensory and motor areas
of human cerebral cortex as determined by direct
recording of evoked potentials and electrical
stimulation. J Neurosurg 51, 476
506.
Wu Q, Garcia-Larrea L, Mertens P et al. (1999)
Hyperalgesia with reduced laser evoked potentials
in neuropathic pain. Pain 80, 209
214.
Wycis HT, Spiegel EA (1962) Long-range results in
the treatment of intractable pain by stereotaxic
midbrain surgery. J Neurosurg 19, 101
107.

References
Yamamoto T, Katayama Y, Hirayama T, Tsubokawa
T (1997) Pharmacological classification of central
post-stroke pain: comparison with the results of
chronic motor cortex stimulation therapy. Pain
72, 5
12.
Yamamoto T, Katayama Y, Fukaya C et al. (2000)
Thalamotomy caused by cardioversion in a patient
treated with deep brain stimulation. Stereotact
Funct Neurosurg 74, 73
82.
Yamashiro K, Iwayama K, Karihara M et al. (1991)
Neurons with epileptiform discharge in the central
nervous system and chronic pain. Experimental
and clinical investigations. Acta Neurochir (Wien)
Suppl 52, 130
132.
Yen HL, Chan W (2003) An East
West approach to
the management of central post-stroke pain.
Cerebrovasc Dis 16, 27
30.
Yoshii N, Mizokami T, Ushikubo T, Kuramitsu T,
Fukuda S (1980) Long-term follow-up study after
pulvinotomy for intractable pain. Appl Neurophysiol 43, 128
132.
Young GB, Blume WT (1983) Painful epileptic
seizures. Brain 106, 537
554.
Young RF, Goodman SJ (1979) Dorsal spinal cord
stimulation in the treatment of multiple sclerosis.
Neurosurgery 5, 225
230.
Young RF, Chambi I (1987) Pain relief by electrical
stimulation of the periaqueductal and periventricular gray matter. Evidence for a non-opioid
mechanism. J Neurosurg 66, 364
371.
Young RF (1989) Brain stimulation. In Wall PD,
Melzack R, eds., Textbook of Pain. New York:
Churchill Livingstone, pp. 925
929.
Young RF (1990) Clinical experience with radiofrequency and laser DREZ lesions. J Neurosurg 72,
715
720.
Young RF, Tronnier V, Rinaldi PC (1992) Chronic
stimulation of the Kolliker-Fuse nucleus region for
relief of intractable pain in humans. J Neurosurg
76, 979
985.
Young RF, Vermeulen SS, Grimm P et al. (1995)
Gamma knife thalamotomy for the treatment
of persistent pain. Stereotact Funct Neurosurg
64(Suppl. 1), 172
181.
Young RF (1995) Commentary. In Schmideck HH,
Sweet WH, eds., Operative Neurosurgical Techniques. Indications, Methods and Results, 3rd edn.
Philadelphia, PA: WB Saunders, pp. 1399
1401.
Young RF, Rinaldi PC (1997) Brain stimulation.
In North RB, Levy RM, eds., Neurosurgical
Management of Pain. New York: Springer,
pp. 283
301.
Young RF (1998) Deep brain stimulation for failed
back syndrome. In Gildenberg PL, Tasker RR, eds.,

References
Textbook of Stereotactic and Functional
Neurosurgery. New York: McGraw-Hill,
pp. 1621
1626.
Zachariah SB, Borges EF, Varghese R, Cruz AR, Ross
GS (1994) Positive response to oral divalproex
sodium (Depakote) in patients with spasticity and
pain. Am J Med Sci 308, 38
40.
Zeki S, Ffychte DH (1998) The Riddoch syndrome:
insights into the neurobiology of conscious vision.
Brain 121, 25
45.
Zimmermann M (1979) Peripheral and central
nervous mechanisms of nociception, pain, and
pain therapy: facts and hypotheses. Adv Pain Res
Ther 3, 3
32.
Zimmermann M (1991) Central nervous mechanisms modulating pain-related information: do
they become deficient after lesions of the peripheral or central nervous system? In Casey KL, ed.,
Pain and Central Nervous System Disease. The

377
Central Pain Syndromes. New York: Raven Press,
pp. 183
200.
Zubieta J-K, Dannals RF, Frost JJ (1999) Gender and
age influences on human brain mu-opioid receptor binding measured by PET. Am J Psychiat
156, 842
848.
Zubieta J-K, Smith YR, Bueller JA et al. (2001)
Regional mu opioid receptor regulation of sensory
and affective dimensions of pain. Science 293,
311
315.
Zuelch KJ (1960) Schmerzbefunde nach operativen
Eingriffen am Zentralnervensystem. Acta Neurochir (Wien) 8, 282
286.
Zuelch KJ, Schmid EE (1953) Uber die Schmerzarten
und den Begriff der Hyperpathie. Acta Neuroveg
7, 147
159.
Zylicz Z (1997) Opioid responsive central pain of
cerebrovascular origin: a case report. Palliat Med
11, 495
496.

INDEX

Above-level pain 115

Acupuncture 234
Adenosine 180, 181, 211, 228, 236
After-sensations 55
Aggravation, drug 182
Alfentanil 151, 174
Allachesthesia 111
Allochiria 111, 119, 321
Allodynia 4, 54, 56, 98, 105, 121, 129, 131, 132, 139,
178, 181, 198, 211, 237, 2427, 336
Amantadine 178
Amitriptyline 179, 180
AMPA (alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid) 177, 337
Anesthesia dolorosa 1, 300
Anterior spinal artery syndrome 327
Anterolateral cordotomies (spinothalamic
tractotomies) 102, 279, 298
Antidepressant drugs 179
Antiglutamatergic agents 178
Arterovenous malformations (AVM) 187, 189, 208
At-level pain 114, 117, 119, 135, 218, 229, 331
Atophanyl 171
Baclofen 52, 121, 138, 175, 182, 208, 22531, 235, 236,
323
Barbiturates 176
Below-level pain 113, 114, 117, 119, 22830, 315, 316,
331, 334
Benzodiazepines 176, 323
Bilaterality 32830
Brain central pain 1, 9112, 116, 119, 121, 128, 178,
181, 254, 255
age of onset 28, 29
causative lesions 9, 10, 102
clinical features 56
coping strategies 232
evoked pains 54, 243
incidence 9, 22, 28
microrecording/stimulation studies 254
pain components 49
pain descriptors 47
pain distribution 40
pain intensity 49
pain quality 47, 49

prevalence 22, 27
psychophysical tests 7097
sensory impairment 5865
sex distribution 28, 31
side of the lesion 40, 41
size of the causative lesions 45
somatosensory findings 56, 64, 6897, 99
sympathetic and other signs and symptoms 99
time to pain onset 28, 35
Brainstem
results of procedures 4, 277, 297
Bulbar tractotomy 102, 106
Bulbar trigeminal tractotomy 111
Buprenorphine 224, 228
Burning pain 59, 84, 100, 101, 117, 120, 129, 132, 218,
298300, 319, 330, 331
Bursting 211, 253, 2656

C fibers 319, 324, 325

Calcitonin 151, 181
Callosectomy 276
Cannabis 146, 148, 174
Carbamazepine 121, 138, 140, 176, 179, 181
Cauda equina lesions 115, 135, 330
Cavernoma 10, 52, 114, 306, 328
Central dysesthesias/paresthesias 1, 7
Central neurogenic pruritus 1, 2, 47, 50
Central pain
diagnosis 5, 1337
examination protocol 136
genesis 6, 7, 239, 249, 265, 315
iatrogenic causes 102, 247
pain types 48, 49, 120
prevalence of 22, 113
somatosensory deficits 56, 121
sudden disappearance of 3005
Central pain diagnosis
bedside examination 133
LEPs 1345
neuroimaging 134
QST 98, 1301, 134
SPECT/PET and pharmacological dissection
1356, 23949
Central pain-allied conditions 2

379

380

Index
Central post stroke pain (CPSP) 1, 9, 2840, 45, 46,
54, 56, 100, 131, 145, 14955, 158, 1725,
18594, 196, 198, 200, 201, 2047, 212, 213, 216,
2208, 239, 240, 2428, 251, 255, 256, 259,
2624, 27681, 28690, 298, 301, 303, 306, 319,
328, 329, 337
Cerebral blood flow (CBF) 207, 208, 2414, 249, 251, 310
Cerebral cortex 240, 241, 301, 308, 312, 322, 334
Cingulate cortex 207, 242, 293
Cingulotomy 291, 293
Citalopram 141
Clonazepam 121, 175, 176
Clonidine 174, 224, 226, 227, 229, 231, 232, 235, 236
Commissural myelotomy 279
Commissurotomies 299
Computed tomography (CT) 45, 136, 3013
Coping 233
Cord central pain 1, 11332, 135, 139, 173, 175,
208, 21217, 2228, 234, 242, 256, 306,
3306
age of onset 116
causes 11314
clinical features 11921, 135
course 128
evoked pains 120, 134
incidence 113
lesion level 113
microrecording; stimulation studies 2567
pain components 120
pain distribution 118
pain intensity 120
pain quality 119, 134
prevalence 113
sex distribution 116
somatosensory findings 121
sympathetic and other signs and symptoms 128
time to pain onset 116
Cordectomies 299, 3302
Cordotomies 299
anterolateral 298
posterior 299
Cortectomy 27694
Deactivations 252, 253
Deafferentation 1, 2, 308, 320, 323, 326, 329
Deafferentation pain 1, 2, 308
Deep brain stimulation (DBS) 199
effects on PAG-PVG 199
effects on Vc 206
efficacy of 208
PAG-PVG 199204, 20810
Vc 199211
Deep pain 136, 152, 241, 245, 255, 256, 276, 295,
300
Descending inhibitory systems 102, 318, 321
Dextromethorphan 141, 142, 147
Diazepam 52, 152
Disappearance of central pain, sudden 3006
Disinhibition 316, 31821, 324, 325, 337
Distraction 233
Divalproex sodium 163
Dorsal column 218, 316
Dorsal horn 2, 3, 196, 218, 237, 256, 309, 320, 321,
332
Dorsal root entry zone (DREZ) 284
coagulation 297

Double lesion syndrome 115

Dronabinol 146
Drug therapy 138
antidepressants 179, 180, 235
cannabinoids 180
combinations of 181
drugs aggravation of CP 182
GABA drugs 139
miscellaneous agents 181
opioids 180, 278, 315, 337
oral, controlled studies 140
oral, uncontrolled studies 160
parenteral, controlled studies 149
parenteral, uncontrolled studies 171
Dynamic reverberation 30738
Dysesthesia 1, 2, 8, 37, 38, 47, 4955, 102, 104,
111, 11315, 117, 118, 129, 1327, 156, 212,
237, 246, 258, 2768, 295, 297, 298, 305,
31618, 331, 336
Ectopic pacemakers 330, 334
Electroconvulsive therapy (ECT) 221, 223
Electroencephalogram (EEG) 252, 255
End-zone pain 113, 115, 119
Epilepsy, central pain and 100, 132
Evoked pain 2, 49, 54, 55, 99, 120, 121, 129, 134, 150,
1769
Fampridine-SR 153
Fentanyl 152, 174
Flexion nociceptive reflex (RIII) 135, 197, 242
Fluoxetine 138, 180
Fluvoxamine 162
Frontal areas 198, 247, 249
Functional imaging studies 8, 134, 23954
Functional magnetic resonance imaging (fMRI)
2446, 250, 327
GABA (gamma aminobutyric acid) 139, 176,
179, 181, 182, 184, 198, 207, 211, 221,
232, 236, 248, 251306, 31113, 3216,
338
GABA responsive patients (class A), 176
GABA unresponsive patients (class B), 176
Gabapentin 53, 138, 144, 146, 1757, 181, 235,
Gasserian ganglion stimulation 221
Genetic predisposition to pain 315
Gigantocellular reticular nucleus 111
Girdle pain 110, 111, 135
Gliosis 333
Glutamate 177, 179, 253, 313, 323
Guanethidine 28890

177,
231,
329,

243

Hemialgic thalamic syndrome 6, 7

Hemispherectomy 102, 103, 328
Historical review 38, 183
Hyperalgesia 49, 105, 107, 109, 115, 139, 180, 265,
328
Hyperpathia 54, 98, 103, 105, 106, 10911, 136, 180,
243, 256, 260, 262, 330, 337
Hypoesthesia 56
Hypophysectomy 294
Hypophysectomy-hypothalamotomy, results of 276
Hypothalamotomy 294
Hypothalamus 221, 294, 312

Index
Iatrogenic lesions 102, 113
Imipramine 173
Incomplete spinal lesions, 120
Insula 196, 207, 233, 2438, 250, 252, 292, 293, 303,
319, 337
Intermittent pain, 120, 330
Intralaminar nuclei 296
Iomazenil. 248
Irritation 7, 334, 335
Itch 12, 503, 265
Ketamine 141, 150, 151, 156, 158, 1735, 177, 178,
181, 182, 184, 186, 187, 190, 253, 309, 323
Koelliker-Fuse nucleus 203
Lamotrigine 142, 166, 175, 177, 178, 235
Lancinating pain 47, 4954, 129, 132, 134
Laser-evoked potentials (LEPs) 134, 135, 197
Lemniscal system, 7, 8, 31618
Lenticular and caudate nuclei, lesions of 104
Levorphanol 145
Lidocaine 53, 153, 156, 171, 172, 174, 175, 178, 190,
2246, 231, 236, 243, 288, 326, 337
Lissauers tract
section of 110
Lobotomies 276
Local anesthetics as diagnostic tool 134
Locus coeruleus/subcoeruleus 208
Magnetic resonance imaging (MRI) 134
Magnetic resonance spectroscopy (MRS) 247
Magnetoencephalography (MEG) 245, 250
Memantine 178, 232
Mesencephalotomies
results of 277
Methadone 180
Mexiletine 141, 153, 172, 174, 178, 179, 181, 235
Motor cortex/MI 184, 185, 187, 1959, 2413, 245,
256, 267, 276, 303, 327, 329
Midazolam 174, 176, 2302, 235
Mirror pain 321
Monoamine oxidase (MAO) inhibitors 180
Morphine, see Opioids
Motor cortex stimulation (MCS) 18499, 223, 242,
256, 276, 303, 327, 328
mechanism of action 195, 218
results 185
Multiple sclerosis (MS) 47, 1302, 179, 180, 239, 260,
298, 306
Myelotomy 109, 279, 297, 298
Naloxone 149, 172, 181, 199
Neural injury pain 2
Neural networks 313
Neuroablation, results of 276
Neuroimaging in CP 8, 134, 23954
Neuromatrix 239
Neurometabolic studies 196, 207, 240, 249
biochemical changes 247
CBF 196, 197, 207, 208, 2414, 249, 251, 310
Neuromodulation 183
chemical 223
drug administration, intrathecal or epidural 224
electrical 183
Neuropathic pain (NP) 1

381
Neurophysiology
microrecording/stimulation studies 25466
Neuroplasticity 249, 324, 326, 327
N-methyl-D-aspartate (NMDA) receptors and
antagonists 17780, 254, 266, 311, 312, 314, 323,
337
Nociceptors 250, 320
Norepinephrine 180, 266, 307, 320
Opioids 115, 138, 155, 1715, 1801, 2312, 235,
315, 337
Overreaction 6
Pain distribution 40, 118
Pain intensity 49, 120
Paraplegia, see Cord central pain
Paresthesia 1
Parietal cortectomy 103, 276
Parietal cortex 2, 28, 237, 242, 244, 247, 248, 276, 292,
304, 307
Parietal cortex stimulation (PCS) 184, 190, 195
Parkinsons disease (PD) 2, 101
Pathophysiology of central pain 30738
Pentobarbital 171
Pentothal 171
Periaqueductal grey matter (PAG) 196, 199, 200, 206,
20810, 233, 297
Peripheral blocks 287
Peripheral neuropathic pain (PNP) 1, 238
Periventricular grey matter (PVG) 199204, 206, 207,
209, 210, 223, 235, 262, 337
Phenytoin 50, 138, 179
Placebo 195, 209, 221, 231, 233, 244
Plasticity 3237, 329
Polysynaptic pathways 8
Positron emission tomography (PET), see
Neurometabolic studies
Posterior columns 99, 109, 317, 318, 3335
Posterior cordotomies 109
Posterior poliotomy 285
Pre- and postcentral gyrectomies, results of 276
Prefrontal cortex 244, 246, 291, 319
Pregabalin 177
Propofol 52, 135, 139, 151, 155, 176, 182, 184, 190,
230, 235, 253, 309, 323, 324
Pruritus, see Itch
Pseudothalamic pain 1
Psychiatric surgery 293
results of 276
Pulvinar 104, 294, 296, 307
Pulvinotomy 2956
Quality of pain 47, 119
Quantitative sensory tests (QST) 2, 98, 130, 131, 134,
142
Radiation of pain 55
Radicular pain 330
Reboxetine 180
Referred pain 112, 321
Remote pain 113
Reticular formation 102, 105, 106, 259, 265, 277, 278,
297, 308, 318, 329, 330, 332
Right thalamic lesions 320
Riluzole 179

382

Index
Selective Serotonin Reuptake Inhibitors (SSRIs) 53,
180
Sensory cortex stimulation (SCS), see Parietal cortex
stimulation (PCS)
Sensory epilepsy 100
Sensory evoked potentials (SEPs) 134, 135, 197, 266,
276
Serotonin 180, 221
Shooting pain 5, 47, 49, 54, 101, 11921, 136, 247,
280, 283, 284, 287, 298300, 330, 331
SI 27692, 307, 309, 311, 312, 325, 329, 337
Single photon emission computed tomography
(SPECT), see Neurometabolic studies
Sodium channel blockers 178
Somatosensory abnormalities 56, 68, 121
Somatosensory areas 249, 252, 27692
Somatosensory evoked potentials (SSEPs) 134, 135,
335
Somatotopic reorganization 258, 262, 327
Spinal cord injury (SCI) 11332, 3306
Spinal cord stimulation (SCS)
efficacy of 21118
mechanism of action 211
results 212
Spinal epilepsy 132
Spinal rhizotomies 300
Spinothalamic fibers 6, 129
Spinothalamic tract (STT) 2, 307, 309, 31521, 331,
332
Spontaneous pain 6, 54, 99, 109, 120, 129, 130, 134,
146, 153, 156, 175, 188, 243, 244, 246, 247, 259,
298, 303, 319, 335, 337, 338
Sprouting 325
Stereotactic C1 midline myelotomy 279, 297
Striopallidal system 102
Summation 7, 55
Supersensitivity 324, 326
Sympathectomies and sympathetic blocks 287, 300
Sympathetically maintained pain (SMP) 300
Syndrome thalamique 5
Syringomyelia 117, 121, 12930, 3347
Tactile sensibility 49, 646, 689
Temporal summation 55
TENS 218, 221
efficacy 220
mechanism of action 218
Thalamic pain, see Central pain
Thalamic stimulation, see Deep brain stimulation
(DBS)
Thalamic stroke 1011

Thalamic syndrome 57
Thalamocortical rhythmicity 311
Thalamoparietal radiations, destruction of 103, 303
Thalamotomies, 2726
Theories of central pain 5, 7, 2379, 249, 276, 293,
310, 31521
small world networking 314
attractors 314
dynamic reverberation 30710
GABA role dissection 326
imbalance/disinhibition 316
neural networks 313
nonlinear dynamics 314
persistent oscillation 313
synchrony 198, 310, 312
thalamocortical rhythmicity 311
thermosensory disinhibition 319
Therapy, alternative approaches 234
Thermal allodynia 303, 336
Thermal sensibility 56, 659, 130, 319, 337
Thiopental 174, 176
Thyamilal 172
Tiagabine 53, 177
Topectomies 103, 293
Topiramate 53, 144, 177, 179
Tramadol 138
Transcutaneous electrical nerve stimulation (TENS)
53, 214, 216, 21821, 235, 316
Trazodone 138
Treatment of central pain
TANG guidelines 235
Tricyclic antidepressants 179
Trigeminal rhizotomy 287
Vagal nerve stimulation 221
Valproate 138, 140, 175, 177, 179
Vc 46, 9, 199211, 255, 25876, 294, 295, 307,
3113
Vcpc, see Vc
Venlafaxine 180
Ventroposterolateral nucleus (VPL), see Vc
Ventroposteromedial nucleus (VPM), see Vc
Vigabatrin 177
Visceral pain 48, 68, 101, 115, 130, 188, 334
Visceral stimulation 66, 68, 115
Wallenbergs syndrome 3, 9, 47, 99, 135
Wind-up pain 55
Ziconotide 229, 231
Zonisamide 53, 179